artificial intelligence and economic growth essay

Artificial Intelligence and Economic Development: An Evolutionary Investigation and Systematic Review

• Published: 11 March 2023
• Volume 15 , pages 1736–1770, ( 2024 )

Cite this article

• Yong Qin   ORCID: orcid.org/0000-0002-4966-7899 1 ,
• Zeshui Xu   ORCID: orcid.org/0000-0003-3547-2908 1 ,
• Xinxin Wang   ORCID: orcid.org/0000-0002-4255-5106 1 &
• Marinko Skare   ORCID: orcid.org/0000-0001-6426-3692 2  

20k Accesses

16 Citations

3 Altmetric

Explore all metrics

In today’s environment of the rapid rise of artificial intelligence (AI), debate continues about whether it has beneficial effects on economic development. However, there is only a fragmented perception of what role and place AI technology actually plays in economic development (ED). In this paper, we pioneer the research by focusing our detective work and discussion on the intersection of AI and economic development. Specifically, we adopt a two-step methodology. At the first step, we analyze 2211 documents in the AI&ED field using the bibliometric tool Bibliometrix, presenting the internal structure and external characteristics of the field through different metrics and algorithms. In the second step, a qualitative content analysis of clusters calculated from the bibliographic coupling algorithm is conducted, detailing the content directions of recently distributed topics in the AI&ED field from different perspectives. The results of the bibliometric analysis suggest that the number of publications in the field has grown exponentially in recent years, and the most relevant source is the “Sustainability” journal. In addition, deep learning and data mining-related research are the key directions for the future. On the whole, scholars dedicated to the field have developed close cooperation and communication across the board. On the other hand, the content analysis demonstrates that most of the research is centered on the five facets of intelligent decision-making, social governance, labor and capital, Industry 4.0, and innovation. The results provide a forward-looking guide for scholars to grasp the current state and potential knowledge gaps in the AI&ED field.

Similar content being viewed by others

Artificial intelligence in the field of economics

Connecting brain and heart: artificial intelligence for sustainable development

Socioeconomic indicators and innovative countries associated with data mining evidence

Explore related subjects.

• Artificial Intelligence

Avoid common mistakes on your manuscript.

Introduction

In recent years, the sound of artificial intelligence (AI) has always been in everyone’s ears, and it seems to be telling us that the arrival of AI is the destiny of the age (Makridakis, 2017 ). Indeed, AI technology is appearing in various forms at all levels of our contact with society, from small daily chatting intelligent robots to large industry and government-level assisted offices, and is quietly changing the way of life around the world (Li et al., 2017 ). By convention, AI is described as a sub-discipline of computer science dedicated to the development of data processing systems and the execution of functions that match human intelligence, such as learning, reasoning, and self-improvement (Peres et al., 2020 ). According to Trifan and Buzatu ( 2020 ), AI is machine learning, that is, a neural network trained on a data set. Drive resources, data resources and computational theory are the three core elements that influence the development of AI. In contrast to any of the technologies that have emerged in the past, AI can get more brilliant at a particular practical task with the accumulation of time owing to its unique learning ability. AI is designed to serve humans in making the best decisions. To this end, AI has been incorporated into operating systems in the hope of creating systems that can assist humans or even be utterly AI-driven in their decision-making (Gomes et al., 2020 ). Progressively, AI is becoming indispensable technological support for daily social life and economic activities (Naimi-Sadigh et al., 2021 ). Its tremendous contribution to sustainable economic development in all industries is rapidly becoming evident, leading it to become an instant focus of attention at the industry, academic and even government levels (Heylighen, 2017 ). Arguably, AI-related activities will be the driving force for further economic development and result in fundamental shifts in the structure and approach to production, and in the quantity and quality of consumption (Vyshnevskyi et al., 2019 ).

However, while people are cheering this inspiring fact, some are expressing their skepticism. Although the widespread application of AI will cause a short-lived economic boost at this stage (Goertzel et al., 2017 ), in the long run, people’s over-reliance on AI is likely to pose some potential threats (McClure, 2017 ). Such as the unemployment fiasco, moral and ethical risks, and personal privacy concerns that are often mentioned by scholars in the literature (Kak, 2018 ). What is more, the technical bottlenecks in the development of AI technology itself also lead to a large gap between the conception of theoretical research and the blueprint in actual practice. In light of recent events between AI and economic development (AI&ED), it is becoming extremely difficult to ignore the existence of the two colliding with each other. Accordingly, a considerable amount of literature has been published on AI&ED. These studies over the past two decades have provided important information on discussions between AI and economic development. More importantly, the evidence shows the increasing urgency and depth of the intersection between AI and various sectors of economic activity. For instance, to allow the power sector to provide good services at competitive prices, Hernández-Callejo et al. ( 2013 ) designed an architecture model for power load forecasting based on artificial neural networks that conduct short-term load forecasting.

The growing breadth and fragmentation of topics at the intersection of AI and economic activity have made it increasingly difficult for scholars to attempt a comprehensive understanding of the field. To make matters worse, the complexity of the topic has led to a diversity of insights, generating a wealth of ideas and investigations on the link between AI and economic development. While there have been some reviews of the literature on AI and economic development, the multifaceted nature of the field suggests that this is still far from sufficient (Aghion et al., 2018 ). On the one hand, short-term studies such as these do not necessarily show subtle changes over time. On the other hand, the available reviews are selective in the literature they employ and the range is usually limited to fit the volume and variety of relevant literature. At the same time, it is not easy for scholars themselves to objectively summarize and sort out the literature (Lee & Lim, 2021 ).

In moving forward to redress this challenge, this paper attempts, through a combination manner of bibliometric analysis and literature review, to gain a one-stop overview on the publications’ performance, collaboration patterns and intellectual structure of the AI&ED domain. More pertinently, this study responds to this practical need by answering the following three broad research questions: (1) What is the performance and current status of AI in economic activities and its related fields? (2) Which research themes in the field of AI&ED have received sufficient attention and exploration in recent years in the existing knowledge? (3) Which research agenda should endeavor in this domain in the future? By doing so, we establish an overview of the basic information in the field of AI&ED and its current status and trends, so as to summarize possible knowledge gaps, provide new ideas for investigation and locate areas of expected contribution for subsequent research (Donthu et al., 2021 ).

The contributions of this study are twofold. Firstly, we position the research perspective at the intersection of AI and economic development. Compared with other investigations, the work in this paper is more contemporary and novel. It helps to establish an understanding of the complexity and interdisciplinary nature of research on the application of AI in economic development. Secondly, the two-phase methodology, i.e., the two-pronged approach of bibliometric analysis and content survey, guarantees the comprehensiveness and reliability of the study (Qin et al., 2022 ). Using advanced bibliometric techniques, the outline of the evolution and knowledge structure of the AI&ED field is outlined. Also, the emerging research on AI applied to economic activities is clearly perceived, which helps theory and practice to go hand in hand. In particular, for the different knowledge streams, we deploy qualitative content analysis to discuss key publications to determine which topics and issues are front and center in the context of AI and economic development, and how the different topics are bundled in the knowledge streams.

In the remainder of this paper, we present how the two-step methodology works in the “ Research Design: A Two-Step Methodology ” section. Based on bibliometric techniques, the “ Results of Bibliometric Analysis ” section is developed from two dimensions: performance analysis and science mapping. In the “ AI and Economic Development ” section, we conduct a systematic literature review of the five themes identified. The “ Discussions and Implications ” section gives discussions and implications. Concluding remarks and limitations end the paper.

Research Design: A Two-Step Methodology

We adopt a two-step methodology to achieve a deeper understanding of the intellectual landscape of the AI&ED research field and the multi-level connections between AI and economic development. The former employs bibliometric techniques to scientifically conduct extensive quantitative analysis of relevant publications for preliminary validation of research ideas. Based on the former, the latter uses a structured literature review approach to describe recent popular mainstream topics in AI&ED to identify potential research gaps. The overall two-step research protocol is depicted in Fig.  1 .

The two-step research protocol

Phase 1: Bibliometric Analysis

The implementation of a bibliometric analysis can empower us to identify the dynamic nature of the AI&ED research field (Qin et al., 2021 ). We chose the most popular and authoritative Web of Science (WoS) Core Collection database as the starting point of the project. In line with the approach of most scholars at this phase, we defined the field boundary using a set of keywords that are coherent with the purpose of the study. To ensure that the final search results include as much of the desired literature as possible, broader search strings were initially identified, i.e., TS = (“Artificial Intelligence” OR “Machine learning” OR “Deep learning” OR “Intelligent agents” OR “Neural networks” OR “Data mining” OR “Natural language processing” OR “Pattern recognition”) AND TS = (“Economic development” OR “GDP” OR “Economy”). In parallel, to ensure state-of-the-art of records, purely peer-reviewed academic journal articles were considered for this study. Only the publications with language in English were taken into consideration. Besides, to guarantee the annual property of the data, we limited the search span during the period from 1900 to 2021. The search was carried out in March 2022, and a total of 2522 items matched these constraints and were initially included in this examination. Prior to the quantitative statistical analysis, we manually checked the titles, abstracts and keywords of the identified documents and those irrelevant publications were removed. In the end, 2211 records were created for this investigation.

To fulfill the objectives of the quantitative analysis and visualization of the retrieved documents, we need to adopt some advanced bibliometric tools. Bibliometrics is based on quantitative methods designed to identify, describe, and evaluate published research (Bretas & Alon, 2021 ; Garfield, 1979 ). Its use of scientific mapping and graphical presentation of reproducible statistics reduces the subjective bias of literature reviews on the one hand, and overcomes the limitations of diagnosis and the error-prone nature of manual summarization on the other (Su & Lee, 2010 ; Tariq et al., 2021 ; van Eck & Waltman, 2010 ). Gradually, the ideas and theories of bibliometrics have become an invaluable manner for many scholars to explore and discover new knowledge in academic research (Qin et al., 2020 ; Wang et al., 2020a , b , c ). In response to this trend, many advanced algorithms and sophisticated visual analysis tools have been developed to help scholars quickly perform bibliometric analysis. In this paper, we apply two bibliometric tools Bibliometrix and VOS viewer that are more mature at this stage. Bibliometrix is a powerful open-source tool developed by Aria and Cuccurullo ( 2017 ), which supports a recommended workflow to perform bibliometric analysis aimed at performing comprehensive scientific mapping work. By using this tool, we accomplished almost all the bibliometric parts of this paper, that is, the performance analysis and science mapping analysis of the collected records, including publication trend, most relevant sources, most influential papers and authors, conceptual structure, and intellectual and social structure. As an equally excellent structured analysis software, VOS viewer is more focused on the graphical representation of bibliometric maps (van Eck & Waltman, 2010 ). With the assistance of its bibliographic coupling procedure, this paper achieves an in-depth exploration and examination of the intellectual structure of the core publications of AI&ED.

Phase 2: Literature Review

In the first phase, we utilized bibliometric analysis to provide an objective, but only cursory, understanding of the intrinsic structure and overall extrinsic performance of the AI&ED domain. This macroscopic model of mathematical statistics appears to be more extensive and clearly demonstrates the connections between different attributes, but it does not allow for profound qualitative conclusions to be drawn. In view of this, clusters formed by core knowledge streams in the bibliographic network based on AI&ED publications are reviewed qualitatively and manually in order to summarize the hot spots and gaps in current knowledge on different topics and thus answer specific research questions. Although the traditional process of qualitative literature analysis can be laced with viewer subjectivity, the benefits of this approach are well recognized (Vallaster et al., 2019 ). Besides, as Gaur and Kumar ( 2018 ) stated, it is the combination of content analysis with other methods that facilitates its tremendous potential. Undoubtedly, bibliometrics perfectly matches the traditional content review (Ante et al., 2021 ). The complementary content analysis allows us to identify hot spots and blind spots in the various research tributaries in AI&ED, thus prompting subsequent research directions to be discovered.

Results of Bibliometric Analysis

Performance analysis.

In this part, we adopt several performance indicators in bibliometrics to provide valuable insights into the AI&ED field. Concretely, we focus on the publication trend, most relevant sources and most influential papers and authors.

Publication Trend

The 2211 documents included in the final dataset generate the annual scientific production in the field of AI&ED, as depicted in Fig.  2 . Studies on AI&ED started in 1986, when Yamashiro posted their seminal work in which online secure-economy preventive control of power system was presented based on pattern recognition (Yamashiro, 1986 ). Although research on this issue has received attention from the scientific community since then, the published studies on AI&ED increased dramatically until approximately 2016, especially during the period from 2018 to 2021. The exponential growth pattern of the field in recent years suggests two facts. The extensive application of AI to economic development and relevant areas is a very recent phenomenon. On the flip side, there exists a fierce argument in management research within AI’s role in the achievement of economic development. According to the visible observed trend in Fig.  2 , research on AI&ED is still immature and in the stage of infancy. With the deepening of AI technology, we can expect a great deal of research in the future dedicated to further enhancing domain knowledge on economic research through AI.

Annual scientific production

Most Relevant Sources

Overall, the 2211 selected documents cover 1096 different sources. Figure  3 sets out the international panorama of the top 20 most relevant sources in the AI&ED field. In this case, we could easily find that the top source comes from “Sustainability” with a total number of 61 publications updated to 2021. The “IEEE Access” owns the second rank with 60 publications, followed by “Energies” (39 publications). With the same number of 39 publications, “Expert Systems with Applications” is in the fourth position. In this regard, investigators concerned with the AI&ED topic need to be particularly attentive to these sources. Moreover, Fig.  4 provides the year-wise growth of the top 5 sources over the period 1986–2021. The temporal evolution of these sources demonstrates that the majority of journals are distributed in a growing trend. In particular, “IEEE Access,” “Sustainability,” “Energies,” and “Journal of Intelligent & Fuzzy Systems” become productive during the last lustrum. Instead, “Expert Systems with Applications” exhibits a slower increase trajectory in recent years.

Most relevant sources

Source dynamics

Most Influential Papers and Authors

Citations for an article are regarded as an appropriate manner to measure its influence and authority in the field (Wang et al., 2021a ). Given this backdrop, highly cited documents over the period 1986 to 2021 in the AI&ED field are assessed, and the top 10 cited publications are exhibited in Table 1 . Nevertheless, the total number of citations (TC) per se does not completely determine the quality of an article, and the time factor usually needs to be considered. Thus, the average number of citations received each year (TC/Y) is also generally deployed as an effective metric for an article’s impact.

Table 1 lists the specific TC and TC/Y across the top 10 documents. Also, the other useful information on them is specified. Evidence from Table 1 indicates that more than half of the publications have been cited more than 300 times in total. Besides, two observations could be obtained from this table. In the first place, five of these documents were pressed before the year 2010, and five after 2010. Surprisingly, the article titled “Automated detection of COVID-19 cases using deep neural networks with X-ray images”, published in 2020, earned a whopping 622 citations. The sudden appearance of the novel coronavirus in 2019 has brought a great impact on the life and health of people all over the world. To accurately detect and diagnose potential people suffering from this disease, an automated assisted diagnosis tool named DarkCovidNet based on deep neural networks was developed by Ozturk et al. ( 2020 ). Furthermore, the article called “Brain Intelligence: Go beyond Artificial Intelligence” on the list, despite being published as recently as 2018, has 409 citations (Lu et al., 2018 ). A novel technology concept named brain intelligence was introduced in their work to break through the many limitations of extant AI. Secondly, in terms of research contents, AI technologies have penetrated various areas of the economy concerned, such as finance, energy and machinery, and are increasingly playing an essential role.

A total of 6871 authors participated in the study on the AI&ED domain, and Fig.  5 depicts the top 10 leading authors in the analyzed dataset. In the specific case of productivity, the top five authors, including Li Y, Hele M, Magazzino C, Wang Y, and Zhang Y, produced 17, 16, 14, 11, and 10 articles respectively. In contrast, the last five authors mostly yielded eight articles. In a nutshell, the distribution of research results in this area is somewhat scattered and lacks core leaders.

Most relevant authors

Science Mapping

With respect to the analysis at the science mapping level, a series of bibliometric methods are exploited here to identify the conceptual, intellectual and social structures hidden in AI&ED issues.

Conceptual Structure

Keywords are a high level of abstraction and generalization of an article’s research content, which empower a good way for scholars to discern the research topic and capture potential trends (Wang et al., 2020a , b , c ; Zheng et al., 2016 ).

Several dominant themes are usually shaped in the development of a particular domain within the research. To this end, Bibliometrix provides the strategic diagram function to identify themes in different phases based on the centrality and density ranking. On the basis of co-occurrence analysis for the author’s keywords, the comprehensive strategic diagram of AI&ED research from 1986 to 2021 is constructed as presented in Fig.  6 . As a result, the nutshell overview of the dominant research topics on AI&ED is highlighted. Obviously, the X-axis (centrality) and Y-axis (density) split the two-dimensional space into four different regions (i.e., quadrants). In this setting, four types of themes with different meanings are clearly distinguished (Cobo et al., 2011 ). Centrality gauges the level of inter-cluster interaction, whereas density measures the level of intra-cluster cohesion (Forliano et al., 2021 ). More to the point, themes that fall in the first quadrant (upper-right quadrat) are usually well-developed and are significant in shaping the field of study. They have high centrality and density values and are usually referred to as motor themes. A theme is characterized by low centrality and high-density values, which is positioned in the second quadrant (upper-left quadrat) as a highly-developed and isolated theme. Diametrically opposed to the thematic characteristics of the first quadrant, themes in the third quadrant (bottom-left quadrat) are not only low in centrality but also low in density, with disappearing or emerging themes gathering here. Lastly, basic and transversal themes usually lie in the fourth quadrant (bottom-right quadrat) with high centrality and low-density values (Lam-Gordillo et al., 2020 ). Visible here is that each theme cluster is composed of a number of keywords, and its name is determined by the most frequent keyword. Besides, the higher the frequency of keywords per theme, the larger the area of the circle will be accordingly.

Strategic diagrams of AI&ED research (1986–2021)

Therefore, five prevalent themes are finally identified in the diagram. Research related to “artificial intelligence,” “big data,” and “Internet of things” is aligned to the first quadrant, suggesting research on these topics dominates and profoundly influences other topics in the AI&ED field. The developed but isolated theme in the second quadrant, namely “neural network,” “optimization,” and “energy management,” should be given sufficient attention to breaking down the silos of research. Interestingly, related studies on “machine learning”, “data mining” and “classification” are recognized as disappearing or emerging themes, which to some extent foreshadows future research frontiers. Not surprisingly, the problems about “forecasting” become the general and broadly researched themes. How AI boosts economic development and finding effective paths to it will be a topic of continuous discussion in the future.

Intellectual and Social Structure

After examining the conceptual structure concerning the AI&ED field, the intellectual and social structure would be further revealed in this part. To be specific, we are committed to visualizing co-citation network and country collaboration map in the AI&ED field. Co-citation analysis is used for the analysis of the cited sources, which allows us to quickly capture the mainstream source communities. In the same way, Fig.  7 outlines the three source clusters amongst the 50 most influential sources. In the first cluster (shown in red), 16 sources are detected, and high-quality journals such as “Neurocomputing,” “Expert Systems with Applications,” and “Decision Support Systems” occupy the main position. 18 sources make up the largest Cluster 2 (shown in blue), in which the representative sources include “Applied Energy,” “Renewable and Sustainable Energy Reviews,” “Energy,” and so forth. In the last cluster (shown in green), 16 sources are more dispersed in the figure, with “Nature” and “Science” journals occupying the center of the diagram.

Co-citation network of cited sources in the AI&ED field

With consideration to the prevalence of cooperation and linkages between authors from different regions or countries, we conduct a collaboration-based assessment of international cooperation. By performing the Collaboration WorldMap function in the Bibliometrix and setting the minimum edges as three, Fig.  8 sheds light on the social structure within the AI&ED domain. Overall, there are 627 pairs of country/region key cooperation on this map. At the same time, the higher the productivity of a country or region, the darker its color is, while the connection of the lines indicates the presence of collaboration, and the more robust the line, the higher the rate of collaboration. The assessment shows that China, the USA, and India are among the world leaders in terms of individual country or regional contributions with 624, 412, and 210 publications, respectively. Another interesting finding shows that scholars from the USA and China are fostering the strongest collaborations, and they are building strong ties with their counterparts around the world. In fact, the highest rate of collaboration between the USA and Chinese scholars has also been maintained, with a total of 66 co-authored articles. As it clearly appears, there are still several authors from different countries or regions who are not involved in this area of communication and collaboration.

Country collaboration map

AI and Economic Development

Bibliographic coupling occurs when two publications cite a third common publication in their bibliographies (Wang et al., 2021b ). As a similarity measure, it is often used to cluster similar research streams. Obviously, the magnitude of coupling is proportional to the relevance of the research topic and content between publications. The significant difference compared to co-citation analysis is that bibliographic coupling analysis can better identify the distribution of recent research topics and current trends in AI&ED, which can inspire us to ponder about future research (van Oorschot et al., 2018 ). Thus, with the assistance of the VOS viewer tool, Fig.  9 visualizes the coherent bibliographic network of the AI&ED literature to detect similar subject areas, and determines the mindset of core researchers.

Bibliographic network of AI&ED publications

Since the bibliographic network generated by the initial 2211 publications cannot identify the number of controllable and valid clusters, we set some filtering conditions and modulate some parameters to derive the number of clusters that can be analyzed. Expressly, to obtain core insights and capture closely linked research results within each cluster, we eliminate unconnected items to show the largest set of connected items. What is more, in our study, we adjusted the minimum cluster size and set it to 12 instead of the default of 1, which makes the final number of clusters more concentrated. In fact, we have also fine-tuned the final rendering of the graph by changing the repulsion parameter to -1, while leaving the attraction parameter as default. Finally, Fig.  9 generates five highly distinguishable clusters that are given different colors to highlight. In what follows, this paper will review these five relatively independent research streams in detail. The five presented broad research topics are: AI supports intelligent decision-making (" AI supports Intelligent Decision-Making " subsection), AI empowers social governance (" AI Empowers Social Governance " Subsection), AI enhances labor and capital (" AI Enhances Labor and Capital " subsection), AI accelerates Industry 4.0 (" AI Accelerates Industry 4.0 " subsection) and AI fuels innovation (" AI Fuels Innovation " subsection).

AI supports Intelligent Decision-Making

In this cluster, how to use AI techniques to maximize successful decision-making in economic problems becomes the main research focus. Intelligent decision-making could be generally understood as the application of the knowledge representation and thinking process of AI into the decision-making theory, by introducing theories and methods from management, computer science and related disciplines for analysis and comparison, thus providing wise and intelligent aid for managers to make the right decisions (Niu, 2018 ). However, the prerequisites for efficient prediction largely determine the likelihood that intelligent decisions will eventually be realized. Forecasting is based on the historical data of things, through certain scientific means or logical reasoning, to make estimation, speculation and judgment on the future development of its situation, and seek the future development law of things. In recent years, the fact that correct predictions (or forecasts) will lead to successful decisions and thus provide maximum economic benefits has increased the interest in predictive modeling. Indeed, in contrast to traditional econometric techniques, AI technology, with its mighty computing power, has injected new blood into scientific forecasting, providing more feasible ideas and solutions for forecasting technology. Also, it significantly improves the accuracy and reliability of forecasting and provides decision support capabilities for various industries that beyond traditional statistical-based analysis (Binner et al., 2004 ). As a consequence, AI-based predictive algorithms are increasingly being considered in various areas of human economic creation.

Energy is of strategic importance to the development and social welfare of any economy (Cen & Wang, 2019 ). Effective forecasting of energy demand, consumption and prices is directly related to the compatibility between the economy and the environment. For example, Ardakani and Ardehali ( 2014 ) developed an optimal regression and ANN (artificial neural network) model for predicting EEC (electric energy consumption) based on several optimization methods, examined the effects of different historical data types on the accuracy of EEC prediction, and then made long-term predictions for two different types of economies, Iran and the United States, respectively. In order to improve the accuracy of oil market price prediction, Cen and Wang ( 2019 ) used Long Short Term Memory, a representative model of deep learning, to fit crude oil prices. Moreover, swarm intelligence approaches, including artificial bee colony (ABC) and particle swarm optimization (PSO) techniques were introduced to evaluate the electrical energy demand in Turkey (Kıran et al., 2012 ). Also for Turkey, Uzlu et al. ( 2014 ) applied the ANN model and TLBO (teaching–learning-based optimization) algorithm to estimate its energy consumption, which also showed good prediction performance. However, a single model cannot always meet the requirements of time series prediction and fuel consumption variation (Liu et al., 2016 ). In parallel, the fact that energy consumption involves a large number of parameters makes its forecasting a complex and challenging task to carry out. To this end, combining the excellent predictive models available is the most straightforward response, and it has proven to be effective (Li et al., 2018 ). Predicting the interrelationship between energy activities and real economic fluctuations is also further explored by relying on AI algorithms. In different domestic and international environments, varying oil price shock incentives can cause different oil price shocks and have different macroeconomic impacts. In response to this problem, Ju et al. ( 2016 ) proposed an ontology-supported case-based reasoning approach to an incentive-oriented AI early warning system, namely the relationship between oil price shocks and the economy early warning system, for predicting the linkage changes between macroeconomic and oil price shocks in China. Furthermore, the economic dependence between urban development policies and energy efficiency improvement was revealed by building a neural network model (Skiba et al., 2017 ). In addition to the energy sector, other areas involved in economic development are also actively incorporating AI technologies to achieve the best forecasting results, such as the spatial prediction of land subsidence susceptibility (Arabameri et al., 2020 ), the prediction of standardized precipitation evapotranspiration index (Soh et al., 2018 ) and predicting the monthly closing price of major USA indices (Weng et al., 2018 ).

In fact, forecasting can also be considered as the process of filling in the missing information, i.e., using the information already collected to generate information that we do not yet have or that we expect to have. Based on the vast amount of available data, AI technology can quickly and efficiently make diagnoses or judgments to help people make the best decisions in a short period of time, minimizing economic risk at the organizational, industry and national levels. At the end of 2019, the sudden onslaught of the novel coronavirus 2019 not only posed a huge threat to people’s lives and health, but also caused a heavy blow to economic development worldwide. As the epidemic continues to spread around the world, diagnosing infected patients has become one of the urgent tasks to be solved at that time. For this reason, many radiological images have been widely used for the detection of COVID-19. In particular, the integration of AI technology allows the diagnosis of patients with COVID-19 infections at a significant advantage (Tsiknakis et al., 2020 ). For instance, Ozturk et al. ( 2020 ) presented a new model for the automatic detection of COVID-19 using raw chest x-ray images. The model can achieve an accuracy of 98.08% for the classification of binary classes and 87.02% for the classification of multiple classes. It is worth pointing out that despite the widespread use and effectiveness of AI in fraud detection, the emergence of new fraudulent vectors has posed severe challenges to fraud detection in the AI framework (Ryman-Tubb et al., 2018 ). Besides, the boom in the fitness industry in recent years has led to a critical need for scientific and practical instructional programs. In light of this, real-time monitoring and guidance based on exercisers’ daily fitness data, supported by AI technology, has become a trend for future fitness applications (Yong et al., 2018 ).

The development of computers and information technology gave rise to the creation of a decision support system (DSS) in the mid-1970s to help decision-makers improve the level and quality of their decisions. Suffice it to say that the rapid advancement of AI technology has given people a wonderful aspiration for the intelligence of traditional DSS (Pinter et al., 1995 ). Later, DSS was combined with AI and expert system technologies, and the prototype of an intelligent decision support system (IDSS) was outlined, enabling the original system to cope with more complex and uncertain decision scenarios. With this opportunity, IDSS has been widely studied by scholars and involved in many human economic activities. For example, to achieve effective management and rapid response to different customer needs in transportation enterprises, He et al. ( 2014 ) proposed a general framework that integrates intelligent technologies as components into the architecture of service-oriented group decision support system, and skillfully used AI technology to solve the conflict problem in distributed group decision-making. The multi-agent system theory and techniques in AI likewise provide essential insights for the development of DSS. To address the complex issues in agricultural development, Xue et al. ( 2013 ) designed an agent-based regional agricultural economy decision support system (RAEDSS) to simulate and evaluate the impact of policies on rural development under different scenarios. Considering that intelligent decision-making should have the ability to explore and discover uncertain environments, scholars have tried to combine fuzzy logic with IDSS to enhance its knowledge representation and reasoning capabilities. Using fuzzy cognitive maps, Albayrak et al. ( 2021 ) developed an IDSS to achieve high yield of honey. In addition, uncertain production goals are extremely common in production plants, and this uncertainty leads to the invalidation of regular management. In view of this, Rodriguez et al. ( 2020 ) proposed an IDSS for production planning based on machine learning and fuzzy logic to solve the closed-loop supply chain management problem.

AI Empowers Social Governance

At present, AI technology is developing deeply and AI application scenarios are enriching, which then calls out a new governance concept and governance form for society. Overall, the new pattern of AI-powered social governance is in the preliminary exploration stage (Mania, 2022 ). What is certain, however, is that AI technology has been used more widely than ever in recent years. These wide ranges of applications are not only reflected in common daily aspects such as image analysis, face recognition and big data analysis, but also gradually rise to the level of major social rulings and human emotional cognition applications (Coglianese & Lehr, 2017 ; Huang et al., 2019 ). At the city level, the great strength of AI in processing big data has contributed to a major change in the urban fabric, a prospect greatly facilitated by the emerging smart city concept that promotes the combination of sensors and big data through the Internet of things (Allam & Dhunny, 2019 ). The core idea of the smart city emphasizes the underlying support of big data, which requires not only tens of thousands of data, but also the integration of multi-dimensional data. This is well evidenced by the full impact of the COVID-19 pandemic. The recent COVID-19 pandemic has prompted a great deal of thought by many scholars about many vital issues and potential complexities for organizations and societies (Dwivedi et al., 2020 ; Iandolo et al., 2021 ), particularly the controversy over data sharing related to the concept of urban health and safe cities (Allam & Jones, 2020 ). On one side, strengthening standardized protocols to increase data sharing will not only help the efficient development of epidemic prevention and control, but also facilitate the further construction and design of smart cities, as well as lead to better global understanding and management of the same. However, it is undeniable that sharing urban health data has the potential to impact the economy and politics of a country or region. Besides, as AI continues to permeate all aspects of human society, some administrative agencies are attempting to employ intelligent algorithms to improve the intelligence of government governance. On a technical level, this is entirely possible. Hildebrandt ( 2018 ) pointed out that data-driven artificial legal intelligence may be much more successful in predicting the content of positive law. Likewise, profound developments in information technology are changing the way banks work, relying more on reliable quantitative information from online and credit bureaus, contributing to AI-based decision-making (Jakšič & Marinc, 2019 ). Finally, over recent years, AI technology is also quietly changing the face and operation of other social industries such as education (Mehmood et al., 2017 ; Williams, 2019 ), marketing (Rust, 2020 ) and accounting (Moll & Yigitbasioglu, 2019 ), seeking to improve economic efficiency.

However, we should also see that while AI accelerates economic development and promotes social governance to a new level, it brings additional challenges to human society in terms of legal norms, moral ethics and governance guidelines that should not be underestimated. Firstly, as mentioned earlier, big data gives AI enough valuable data to support it. Generally speaking, the larger and more dimensional the data, the more promising the final effect of intelligent algorithms, which inevitably involves individual-level data analysis, collection and application. Scholars have long debated the protection of personal data and concerns related to privacy (Kak, 2018 ). On the one hand, some scholars have called for striking as much of a balance as possible between data protection and data-related concerns (Dwivedi et al., 2020 ). On the other hand, some scholars pointed out that no one owns data and that property rights protection of data is not appropriate to promote better privacy, more innovation or technological progress, but is more likely to stifle freedom of expression, freedom of information and technological progress. Thus, the case for property rights to data is not compelling, and there is no need to create new property rights for data (Determann, 2018 ). Secondly, Allam and Newman ( 2018 ) cautioned against the blind acceptance of technology and encouraged further embedding into the social fabric. Such a reminder stems in large part from the ethical issues of fairness, responsibility or subjectivity that AI can raise. Research in AI could be roughly divided into three stages: mechanical AI, thinking AI and feeling AI. While mechanical AI is already mature and thinking AI is developing rapidly, the highest level of feeling AI is progressing slowly (Huang et al., 2019 ). Since AI at this phase does not possess self-awareness, AI platforms are not neutral technologies, they are designed with a purpose and exhibit bias and human rights violations (Bourne, 2019 ). Additionally, government agencies are beginning to widely adopt AI technology for constitutional democracy and administrative decision-making, and concerns have increased over digital robots replacing the government sector. At the same time, reliance on AI has also led to an increasing challenge to human subjectivity. Therefore, in response to the above challenges, human workers must pay more attention to the extension of the empathy and emotional dimension in their work (Huang et al., 2019 ). On the flip side, creating a new culture that incorporates the principles of democracy, rule of law and human rights through the design of AI as well as considers diversity in the design and implementation of algorithms is a viable solution for the future (Nemitz, 2018 ; Turner Lee, 2018 ).

AI Enhances Labor and Capital

The debate over the conclusion that AI enhances the quality of labor as well as the quality of capital. In the traditional economic development model, the three production factors, i.e., capital, labor and total factor productivity (TFP), determine the development dynamics of the economy. When the two physical factors, capital and labor, rise in quantity or are used more efficiently, they contribute to economic development. Of course, an increase in TFP due to technological or innovative advances would also generate economic development. Collectively, it seems to be an undisputed fact that the widespread use of AI promotes economic development. More importantly, a large body of empirical literature supports this view as well (Chattopadhyay & Rangarajan, 2014 ).

Firstly, the development of AI has significantly reduced the cost of traditional automation while creating an opportunity for the era of intelligent automation (von Joerg & Carlos, 2022 ). Although traditional automation technologies have led to dramatic increases in labor productivity, specific and homogeneous settings, allow them to perform only simple and repetitive tasks. In contrast to the former, the era of intelligent automation has created a new kind of virtual labor force, which can be considered as a new factor of production. This phenomenon, on the one hand, diminishes the dependence on manual labor at the current stage of production and triggers the substitution of capital for labor (Autor, 2015 ). On the other hand, due to its self-learning and self-renewal characteristics, AI will effectively solve the complex labor needs of the many automated jobs in real life (Bahrammirzaee et al., 2011 ). In particular, this change in the structure of production factors will rapidly produce high-end labor, which in turn will significantly boost economic development (Vivarelli, 2014 ).

Secondly, with powerful and innovative AI technologies, the efficiency of the existing capital and labor has been enhanced to an unprecedented degree, while enabling the skills and capabilities of labor and physical capital to be also supplemented and improved. In fact, in addition to the above-mentioned substitution relationships, there are also many complementary relationships between AI and human intelligence (Huin et al., 2003 ). With human–machine collaboration, workers’ productivity can not only be effectively utilized and extended, but also be motivated to focus on the areas they are good at and thus do more creative work. The scenario of human–machine integration has led to increasing labor productivity (Wolff, 2014 ). For example, accurate estimation of the local scour depth concerning bridge piers is crucial for engineering design and management, which places higher demands on the professionalism of bridge engineers. To this end, a new hybrid smart artificial firefly colony algorithm-based support vector regression model was developed to predict the scour depth near bridge piers by Chou and Pham ( 2017 ). The results showed that the model could effectively assist the concerned staff in constructing safe and cost-effective bridge substructures. In terms of improving capital quality, as described in the “AI Supports Intelligent Decision-Making” section, AI is able to model, predict and ultimately optimize decisions in real time from massive amounts of data in the production process. It can almost completely avoid the problems of low accuracy, low integration and low adaptability in production activities, and achieve intelligence in the production process, thus realizing capital efficiency improvements. For the manufacturing industry, this is particularly evident. AI has become an important driver for intelligent manufacturing technology innovation, promoting economic development and improving people’s quality of life. Research results showed that the adoption of highly interconnected and deeply integrated intelligent production lines would lead to significant improvements in manufacturing productivity as well as a corresponding reduction in the number of system instructions (Hu et al., 2018 ).

Finally, AI’s ability to increase TFP across the board is well documented, and some existing studies even categorize it as a new factor of production that will further fuel economic development in the future. Footnote 1 Nevertheless, in the long run, many scholars are divided on the question of whether the progress of AI will play a sustainable role in promoting economic development. The negative school of thought believes that AI will replace labor and take over human jobs, which will likely lead to unemployed people much faster than productivity can be increased (Vermeulen et al., 2018 ). In a situation where the labor market is disrupted, income inequality and mass unemployment among workers are probably creating a further future of high unemployment and even economic stagnation (Frey & Osborne, 2017 ). As a consequence, AI’s boost in the economy is seen as unsustainable (Vermeulen et al., 2018 ). As noted by Gasteiger and Prettner ( 2017 ), human dependence on AI would eventually lead to an economic rout, as the utilization of automation inhibited wage development and thus investment growth. In contrast, the positive school of thought argues that while AI can rapidly replace labor, the AI revolution will not necessarily have a fatal impact on employment. Specifically, on the one hand, the development and application of AI technology still require many human resources for research and development and design, as well as the operation and maintenance of AI equipment cannot be separated from the participation of senior technical personnel. This demand for high-end human capital creates a higher economic value, but of course also puts forward higher requirements for the quality of the future workforce (Chen et al., 2009 ). On the other hand, the skill requirements of jobs are dynamic, and the impact of AI is likely to generate new labor demand and new job opportunities (Frank et al., 2019 ). The historical experience of the industrial revolution tells us that at this stage, human beings are likely to be in a short transitional period with frictional unemployment, followed by economic prosperity (Vermeulen et al., 2018 ). On top of everything else, some observers consider that the impact of AI on economic development and employment depends heavily on institutions and policies, and that inappropriate labor market and education policies may reduce the positive impact of AI and automation on employment (Aghion et al., 2019 ).

AI Accelerates Industry 4.0

According to our consensus, the Industry 1.0 era was marked by the invention of the steam engine by the Englishman Watt, which exponentially increased the efficiency of production technologies that previously relied on human and animal labor. The widespread availability of electricity has inaugurated the era of Industry 2.0. In this context, the productivity of factories has been developed and further improved. The Industry 3.0 era then witnessed the advent of computers and automation (Syam & Sharma, 2018 ). And in 2013, as the German government introduced the concept of Industry 4.0, it instantly attracted the attention of various countries and industrial giants (Carayannis et al., 2022 ). Industry 4.0 can be characterized as the emergence of cyber-physical systems involving entirely new capabilities for people and machines (Mhlanga, 2020 ). Even though these capabilities rely on the previous phase of Industry 3.0, the continued incorporation of extraordinary technologies has allowed for a long optimization of the third computerized industrial revolution (Sharabov & Tsochev, 2020 ). At the same time, the technology embedded in Industry 4.0 has created a new way of human life at this stage. Underpinned by these disruptive technological advances, Industry 4.0 aims to blur the boundaries among the physical, digital and biological worlds (Huynh et al., 2020 ). Simply put, it is expected to establish a highly flexible, personalized and digital production pattern of products and services, where the original industry boundaries will be broken down and the industry chain will be redefined (Sharabov & Tsochev, 2020 ). Journal articles and related reports in the context of Industry 4.0 indicate a huge demand for developing reliable and usable AI for real-world applications (Lee & Lim, 2021 ). It is foreseeable that AI will play an integral role in the future production paradigm of Industry 4.0 (Skrop, 2018 ). There seems to be a consensus among social scientists that AI is the key technology of the fourth industrial revolution (Liu et al., 2021 ).

The principal features of Industry 4.0 are technological transformations, digital revolution and AI (Wang et al., 2020a , b , c ). More precisely, Sanz et al. ( 2021 ) pointed out that intelligent and automated solutions should be included in industrial processes that employ AI (AI-driven framework) to be competitive in the Industry 4.0 paradigm that essentially affects manufacturing. For this reason, a great deal of research has been conducted on how to combine and embed AI into the existing Industry 4.0 manufacturing value chain (Peres et al., 2020 ). To meet Industry 4.0 manufacturing standards, Nasr et al. ( 2020 ) proposed a hybrid adaptive neuro-fuzzy inference system (ANFIS) based on a multi-objective particle swarm optimization approach to obtain optimal combinations of milling parameters and matching rates to minimize feed force, depth force, and surface roughness. Artificial neural network model for dynamic behavior optimization of robotic arms, an AI technology, was designed to improve the sustainability of Industry 4.0 (Azizi, 2020 ). Furthermore, more places for AI in Industry 4.0 have been identified and perceived by researchers, such as predictive analytics, predictive maintenance, industrial robotics, inventory management and computer vision (Sharabov & Tsochev, 2020 ). Collectively, industrial AI excels in five dimensions: infrastructures, data, algorithms, decision-making, and objectives (Peres et al., 2020 ). There is no doubt that the role of AI is central to the factory of the future, driven by the Industry 4.0 vision and reflected in the great blueprint for the factory of the future (Bécue et al., 2021 ). From an industrial perspective, AI can be viewed as enablers for systems to sense their environment, handle the data they acquire and address complicated tasks, as well as study from experience to enhance their ability to tackle particular challenges (Peres et al., 2020 ). While a high degree of autonomy is one of the core requirements for the future of Industry 4.0, the injection of additional human intelligence may be more beneficial to the operation of future factories and remains true, at least from this phase (Peres et al., 2020 ). In this regard, different levels of autonomous systems are more in line with the differentiated needs of factories at this stage.

From the above description, we can get that the basic concept of Industry 4.0 lies in the organic combination of hardware and software devices, so as to build a smart factory where people, machines, and resources communicate and collaborate with each other (Dopico et al., 2016 ). Currently, Industry 4.0 is a common trend in international development, bringing new opportunities to the economic expansion of many countries (Pham-Duc et al., 2021 ). However, it is not an easy task to truly implement the Industry 4.0 framework in industrial manufacturing processes (Sanz et al., 2021 ). The realization of this digital revolution is costly, and it is even sometimes impossible to quantify (Trifan & Buzatu, 2020 ). In fact, people are questioning whether the era of Industry 4.0 will ever exist, because the event space is infinite. And the actual software and hardware will never cover the infinite event spaces (Vogt, 2021 ). Besides, despite the potential of industrial AI, a large amount of training data and a large amount of computing power are required to make it suffer from a very precarious end as well (Sharabov & Tsochev, 2020 ). What is worse, real factory environments provide unique and difficult challenges for which organizations are not ready (Peres et al., 2020 ). And the physical nature of the systems and processes that industrial AI deals with leads to special constraints that other types of AI do not face (Bécue et al., 2021 ). For instance, the dynamics of anomalous and expected behaviors can cause the original fixed settings to be unable to accurately determine the boundaries between them, making it difficult to detect new threats, which can eventually lead to a series of industrial production security problems (Luo et al., 2021 ). Still, AI may have a perfect niche for its flourishing and implementation in industrial environments, as its applications can answer different questions and possibilities in each of the main pillars of the Industry 4.0 construct (Dopico et al., 2016 ).

AI Fuels Innovation

It is now widely accepted that the advent of AI technology has disruptively improved productivity, but its radiating effect of driving innovation through economic diffusion is rarely talked about or even valued. Innovative thinking and creative ideas are becoming mainstream as people slowly get used to the pounding of the fourth industrial revolution era (Chen, 2022 ). Imagine that if imagination is lost, progress may only be a short-lived blessing (Shakir et al., 2019 ). Mechanical improvements or instability have a long history of impacting innovation, as has AI, endowed with human intelligence (Shakir et al., 2019 ). Within a business perspective, innovation is a multi-stage process by which organizations transform ideas into new or improved products, services or processes to successfully move forward, compete and differentiate in the marketplace (Buhmann & Fieseler, 2021 ). Since AI at this stage is primarily characterized by expanding all aspects of human performance, it is not possible to achieve a high degree of autonomy, or even full autonomy, for the time being. We may question whether AI can take up the burden of influencing or even dominating the innovation management process. At first glance, the vision of AI being used to facilitate innovation purposes seems to be nonsensical. After all, the ability to innovate has traditionally been considered a uniquely human survival capability (Haefner et al., 2021 ). So far, decisions in the innovation process have been made by humans. Just imagine what it means when they are replaced by machines (Verganti et al., 2020 ). Nevertheless, along with the gradual blurring of the boundaries between AI and humans, a large number of cases tell us that AI promises to give birth to different explanations and inventions than before. This groundbreaking progress suggests that AI can be defined as the invention of an inventive method. In other words, AI has the ability to increase innovation productivity by helping human innovators with all the supportive tasks that ignite the creative spark and collate innovation propositions based on their merits (Samid, 2021 ). This is particularly evident in business activities. AI becomes a technology driver for business pattern innovation by steering decisions and automating services to leverage business practices that improve efficiency and profitability (Anton et al., 2021 ). Arguably, AI plays the role of creative enabler and partner to innovation managers in their innovation process (Kakatkar et al., 2020 ). More broadly, AI does have the potential to innovate on its own and to disrupt the entire innovation process under conventional perception, thus fundamentally changing the traditional innovation generation patterns (Hutchinson, 2021 ). The same view is shared by Cockburn et al. ( 2018 ). They claimed that AI also has the potential to transform the innovation process itself, with potentially equally far-reaching consequences, and it may dominate the direct impact over time. Moreover, it is not just about improving the efficiency of research activities, but about creating new scripts for innovation itself.

Certainly, the great human emphasis on the adoption of AI technology in the innovation process stems mainly from compromise with the reality of the environment. To begin with, today’s increasingly turbulent and competitive innovation environment has inevitably created extremely difficult survival conditions. In addition, with the exponential increase in the amount of information collected by organizations or companies, limited human resources can no longer demonstrate the confidence to handle the vast amount of information. More importantly, these organizations or companies are no longer willing or even able to bear the high human cost to cope with this challenge and carry out innovation activities (Haefner et al., 2021 ). On the other hand, AI’s revealed strengths on the road to innovation have forged its current brilliance. In particular, in addition to meeting human’s high efficiency in product or service design, AI injects unique sensory experiences into product or service with its powerful humanization, intelligence and experience interaction, immensely satisfying people’s rich spiritual needs (Wang et al., 2019 ). Also, a number of outreach studies are filling and enriching this invigorating field. For the intrinsic mechanism of AI influencing technological innovation, Liu et al. ( 2020 ) gave a possible answer. They argued that it is because AI ultimately facilitates technological innovation by accelerating knowledge creation and technology spillover, improving learning and absorptive capacity, and increasing investment in R&D and talent. Exciting work on measuring the speed of AI innovation was then developed by Tang et al. ( 2020 ). According to their experiment, 5.26 new researchers were entering AI every hour in 2019, more than 175 times faster than in the 1990s. Additionally, the experience of AI to accelerate innovation varies across countries, especially with the dominant discourse currently committed to it in the West (Alami et al., 2020 ).

The point has to be made that while AI brings great benefits to innovative work, it also creates uncertain risks. Since the time when humans are not satisfied with applying AI to products and services, but rather into the innovation process, resulting in new products and new value chains (Davenport & Ronanki, 2018 ). People are slowly realizing that too much reliance on AI may generate a major threat in the near future (Cath, 2018 ). As such, the concept of responsible innovation is presented to address the ethical issue of responsibility at the boundary of innovation in AI, which provides a path for theoretical reflection and realistic response to the innovation and development of AI (de Saille, 2015 ). According to Buhmann and Fieseler ( 2021 ), responsible innovation in AI should be reflected in the following points. Foremost, the responsibility to avoid damage, i.e., risk management methods that should control potential hazards. Secondly, human-centered, the origin of all innovation is to serve human beings. Finally, governance responsibility stands for the responsibility to create and support global governance structures that can facilitate the first two responsibilities.

Discussions and Implications

This work attempts to provide a bibliometric analysis and a methodological study on the scientific knowledge of the performance of publications in the field of AI&ED. For this purpose, a series of activity metrics and precise content analysis are executed. The theoretical contributions and implications for practice are discussed in the following section.

Theoretical Implications and Roadmap for Future Research

As stated previously, AI technology is commonly applied to all levels of national economic development, and its driving role is indisputable and has a broad scope for development. As such, our survey makes several theoretical contributions and insights for further research on AI applications in the economic field.

Firstly, the result of the publication trend shows that the proliferation of AI in the economy domain has been unprecedented in recent years. In particular, the advent of the post-pandemic era has intensified the reliance on and desire for AI for economic development. As a result, future research efforts are foreseeable. However, as far as the publication channel is concerned, the quality of the research is not yet high enough. The percentage of top journals is very low. The intuition behind this embarrassing situation may be on the one hand that the collection, analysis and processing of the underlying economic big data are hindered. On the other hand, the model of AI serving economic development has not yet matured and is still in a lower quality period at this stage. Due to the interdisciplinary nature of this field, it is necessary to strengthen the scenario-based capabilities of the theoretical foundation of AI. In particular, the consolidation of the theoretical foundation will facilitate the construction of a well-functioning paradigm. On this basis, we encourage future scholars to focus on the effectiveness and practicality of interdisciplinary integration to follow the actual needs of economic development.

Secondly, we provide a list of the most cited work as well as a list of the most published authors, as this survey ascertains the most influential works and the most passionately researched scholars. Thus, the hot articles provide a good theoretical cornerstone for exploring substantial breakthroughs in future research. More importantly, practitioners can track their latest work and contributions to gain cutting-edge wisdom and guidance. It is worth noting that the implementation of AI in economic scenarios requires the joint participation of multiple stakeholders and policy makers. Highly productive academics’ ambitions often lack practical experience, a gap that can lead to stalemates and dilemmas in AI execution. Thus, future researchers can connect multiple parties and actively collaborate to address this potential threat. More directly, the addition of a community of practitioners will accelerate the exploration and theoretical advancement of the AI&ED field.

Thirdly, a visual topic distribution map is executed to complement the existing content analysis to present the distribution and focus of current mainstream research topics. For example, the results of the conceptual structure reveal that researchers may have greater interest and passion for topics related to deep learning, data mining and classification in the future. Thus, we advocate more future agendas around these valuable and promising research hotspots for further expansion of the AI&ED field. However, the problem of AI-based prediction in economic activities remains the spotlight of research at this phase. For this reason, how to effectively use a large amount of economic information and improve the systemic as well as scientific aspects of forecasting issues in economic development has become an urgent concern. In addition, economic problems are often filled with a large amount of uncertainty and ambiguity, which is exacerbated by the emerging COVID-19 pandemic (Ozturk et al., 2020 ). Data uncertainty and cognitive uncertainty in forecasting need to be reconsidered. Based on these facts, uncertainty prediction modeling should be highlighted in future AI&ED problems.

Fourthly, we reveal the collaborative associations and social structure at the country/regional level, and the relative contribution of each country/region to AI&ED research is identified. The results show that China is at the top of the world list in terms of contributions from individual countries or regions. In addition, the USA has established the strongest academic ties with China. This initiative is to be encouraged and supported. The generalizability of AI in solving economic problems needs to be proven in different countries/regions. Cross-border research collaboration can improve the applicability and robustness of AI models. For example, comparing differences in energy forecasting and warning mechanisms across countries/regions is an essential subject to be addressed.

Finally, the bibliographic coupling analysis identifies five mainstream knowledge themes and clusters in the recent AI&ED fields of “intelligent decision-making,” “social governance,” “labor and capital,” “Industry 4.0,” and “innovation.” The content analysis traces the research boundaries and trends of the five sub-topics to provide directional guidance for future research. More importantly, scholars can use the findings of this survey to focus on new and less-researched issues to promote deeper adoption of AI in economic development. Specifically, (1) most of the existing articles repeatedly emphasize the superiority and intelligence of AI in economic decision-making scenarios, but the existing level of AI technology is only limited to specific scenarios and settings, forming a single point of breakthrough in the AI field. Currently, AI-based intelligent decision-making systems for economic activities are highly prone to fail under slight changes. Therefore, broadening the extension and stability of intelligent decision-making in economic activities is an important breakthrough in the future. Of course, a general-purpose integrated intelligent system is also one of the directions to consider. (2) Admittedly, AI has greatly enriched and improved the means of human social governance as well as the efficiency of governance. However, risky events in the governance process, such as fairness of judging, the bias of algorithms and the privacy of users, often trigger governance failure. Therefore, we propose to include the construction of a rule of law system and ethical framework for AI in the future research agenda (Turner Lee, 2018 ). (3) AI is a new manifestation of technological development and can be regarded as a complex combination of capital and labor. Therefore, how to prevent the “unemployment panic” caused by the imbalance of labor and capital substitution in the wave of AI reshaping the economic development paradigm is an urgent issue to be solved in the future. Moreover, it is also necessary to establish and improve the appropriate institutions. (4) There is no denying that AI is at the forefront of leading Industry 4.0, enabling the construction of Industry 4.0 to rise to a new level. However, our review shows that the AI-driven Industry 4.0 framework is still in the blueprint planning. How to build out an Industry 4.0 smart factory with practical operations is still the main melody of the future. (5) AI has great potential to enhance human innovation activities, contributing to innovation-driven economic development. However, our survey shows that uncertain risks lie behind the increased efficiency of innovation. To effectively capture and curb the spillover of these risks, the concept of responsible innovation needs to be further refined and implemented in the future (Buhmann & Fieseler, 2021 ).

Practical Implications

Moreover, the findings of this study also help us summarize a number of practical implications for future development with the aim of removing obstacles to the future development path of AI in terms of its widespread acceptance in economic activities.

Firstly, our bibliometric analysis and content review help practitioners gain a comprehensive understanding of the current state of development of AI as an emerging technology in various areas of human economic development. More importantly, practitioners can increase their confidence that AI will change the future landscape of economic activity and gain possible guidance for their practice from this work.

Secondly, practitioners can derive highly condensed findings and research boundaries from bibliometrics to discuss design choices and trade-offs to remove major barriers and obstacles to the inclusion of AI in economic activities. More pertinently, practitioners in the AI industry may benefit from our survey for more nuanced applications and designs.

Thirdly, our social structure analysis identifies the countries/regions that have achieved more results in the AI&ED field. This finding helps practitioners understand where to seek appropriate collaboration opportunities or advice (Zhang et al., 2021 ). Fourthly, our review argues that AI is rapidly becoming the new frontier of competitive differentiation for economic development in countries around the world. To this end, the work can help leaders as well as policymakers to capture the potential of AI.

Finally, our findings suggest that AI needs to focus on the legal and ethical dimensions of its involvement in human economic activities. We, therefore, call on policymakers to pay attention to these factors on the path to deepening the role of AI. For example, the safety risks of AI technology should not be underestimated. The security concerns of AI have been mentioned numerous times in the existing literature, including ethical security, technical security, data security and so on. For the technical level deficiencies, the government should increase the financial investment and policy protection in this area to provide a good external environment for the development of the AI industry. When it comes to data and ethical security, on the one hand, people must be aware of the privacy nature of data itself, and respect human privacy by establishing moderate legal provisions to address data security as well as embody data privacy protection without impeding the development of AI technology. On the other hand, a professional code of ethics for AI should be developed. In the process of AI design and development, human ethical guidelines and humanism are incorporated, and efforts are made to find best practices that make AI decisions more ethical.

Conclusions and Limitations

In this investigation, the aim is to provide a synthesized review of the extant studies that specialize in the application of AI technology in the economy and related fields. Answering this critical issue requires detailed knowledge that overcomes the fragmented feature of scientific debate in this area. As such, we combine advanced bibliometric techniques with a traditional qualitative literature review to balance the quality and quantity of this discussion. Specifically, a total of 2211 articles published in the WoS Core Collection database were collected for bibliometric analysis and literature survey.

Using the bibliometric tool Bibliometrix, we conducted a performance analysis and science mapping analysis of publications to visualize the landscape and evolution of the AI&ED field and to capture the trajectory of themes over time. The intuitive results show that articles on AI&ED have only emerged in recent years, especially in the last three years, and are now the focus of much scholarly interest. In addition, the most relevant publication sources are concentrated in “Sustainability,” “IEEE Access,” “Energies,” and “Expert Systems with Applications”. In particular, “IEEE Access” has seen the most significant increase in the number of publications in this field in the last few years. The most influential paper was published in 2020 by Ozturk et al. ( 2020 ) in “Computers in Biology and Medicine”, entitled “Automated detection of COVID-19 cases using deep neural networks with X-ray images”. As of the time of data collection, this paper has been cited more than 622 times and may be deemed as an extraordinary work in the field. In addition, the scholar Li Y has produced the most articles and is the most active author in the field. The science mapping draws the conceptual, intellectual and social structure across the AI&ED domain. The distribution of topics in the four quadrants and the evolution of topics over time provide a clear picture of the current knowledge structure and orientation of the field. Overall, the extended conversations on the “big data” and “Internet of Things” are still hot topics at this stage. Prediction-related research is an enduring and widely discussed topic in the field. As expected, the agenda for COVID-19 is emerging. More importantly, there is a close international exchange of scholars from different countries/regions working in this field.

On the other hand, with the support of the bibliographic coupling function embedded in the VOS viewer, we identify five key topic areas that are currently the most popular under AI&ED research: AI and intelligent decision-making, AI and social governance, AI and labor and capital, AI and Industry 4.0, and AI and innovation, which is also an outstanding result of this study. In response to these frontier topics, we run a systematic review to gain insight into each economic subfield. For the practitioner sphere, this work provides theoretical basis and guidance to those currently employed in the field, enabling them to quickly seize the unlimited potential of AI in economic development. Nonetheless, for researchers working in this field, we outline the profile of each topic area and the research gaps, which will have an important enlightening force and stimulating effect on future research in this field.

Limitations and Future Scope for Research

Although the study design of this paper ensures the relevance and reliability of the final results, the generalizability of these results is still subject to certain limitations. First and foremost, for the determination of the data sources, we followed the internationally generally observed rule of access based on the WoS core collection platform. Although the quality of the literature was ensured, a small amount of valuable literature was still hidden in other databases, which to some extent makes the integrity of the sample questionable. Therefore, expanding reliable data collection channels in future research is a feasible way to improve our study. Secondly, out of research needs, we only considered most of the literature in the bibliographic coupling network mapping instead of all the literature, which may cause some bias in the final clustering results. To get a broader picture of the research clusters in this field, we encourage future researchers to consider more literature information to obtain more general and delicate insights. Finally, while the utilization of bibliometric techniques in the paper reveals its advantages of comprehensiveness, it also exhibits its shortcomings of not being able to take into account many details. For the time being, we can only rely on the future improvement of the discipline to remedy this deficiency. Notwithstanding these limitations, this work offers valuable insights for the future boom in the AI&ED field.

https://www.accenture.com

Aghion, P., Jones, B. F., & Jones, C. I. (2018). Artificial intelligence and economic growth. In The economics of artificial intelligence: An agenda (pp. 237–282). University of Chicago Press.

Aghion, P., Antonin, C., & Bunel, S. (2019). Artificial intelligence, growth and employment: The role of policy.  Economie et Statistique , 510 (1), 149–164.

Alami, H., Rivard, L., Lehoux, P., Hoffman, S. J., Cadeddu, S. B. M., Savoldelli, M., & Fortin, J.-P. (2020). Artificial intelligence in health care: Laying the Foundation for Responsible, sustainable, and inclusive innovation in low- and middle-income countries. Globalization and Health, 16 (1), 52.

Google Scholar  

Albayrak, A., Duran, F., & Bayir, R. (2021). Development and evaluation of a web-based intelligent decision support system for migratory beekeepers in Turkey to follow nectar resources. Journal of Apicultural Research, 60 (3), 396–404.

Allam, Z., & Dhunny, Z. A. (2019). On big data, artificial intelligence and smart cities. Cities, 89 , 80–91.

Allam, Z., & Jones, D. S. (2020). On the coronavirus (COVID-19) outbreak and the smart city network: Universal data sharing standards coupled with artificial intelligence (AI) to benefit urban health monitoring and management. Healthcare, 8 (1), 46.

Allam, Z., & Newman, P. (2018). Redefining the smart city: Culture, metabolism and governance. Smart Cities, 1 , 4.

Ante, L., Steinmetz, F., & Fiedler, I. (2021). Blockchain and energy: A bibliometric analysis and review. Renewable and Sustainable Energy Reviews, 137 , 110597.

Anton, E., Oesterreich, T. D., Schuir, J., Protz, L., & Teuteberg, F. (2021). A business model taxonomy for start-ups in the electric power industry - The electrifying effect of artificial intelligence onbusiness model innovation. International Journal of Innovation and Technology Management, 18 (03). https://doi.org/10.1142/s0219877021500048

Arabameri, A., Saha, S., Roy, J., Tiefenbacher, J. P., Cerda, A., Biggs, T., & Collins, A. L. (2020). A novel ensemble computational intelligence approach for the spatial prediction of land subsidence susceptibility. Science of the Total Environment, 726 , 138595.

Ardakani, F. J., & Ardehali, M. M. (2014). Long-term electrical energy consumption forecasting for developing and developed economies based on different optimized models and historical data types. Energy, 65 , 452–461.

Aria, M., & Cuccurullo, C. (2017). Bibliometrix: An R-tool for comprehensive science mapping analysis. Journal of Informetrics, 11 (4), 959–975.

Autor, D. H. (2015). Why are there still so many jobs? The history and future of workplace automation. Journal of Economic Perspectives, 29 (3), 3–30.

Azizi, A. (2020). Applications of artificial intelligence techniques to enhance sustainability of Industry 4.0: Design of an artificial neural network model as dynamic behavior optimizer of robotic arms. Complexity, 2020 , 8564140.

Bahrammirzaee, A., Ghatari, A., Ahmadi, P., & Madani, K. (2011). Hybrid credit ranking intelligent system using expert system and artificial neural networks. Applied Intelligence, 34 , 28–46.

Bécue, A., Praça, I., & Gama, J. (2021). Artificial intelligence, cyber-threats and Industry 4.0: challenges and opportunities. Artificial Intelligence Review, 54 (5), 3849–3886.

Binner, J. M., Gazely, A. M., Chen, S.-H., & Chie, B.-T. (2004). Financial innovation and divisia money in Taiwan: Comparative evidence from neural network and vector error-correction forecasting models. Contemporary Economic Policy, 22 (2), 213–224.

Bourne, C. (2019). AI cheerleaders: Public relations, neoliberalism and artificial intelligence. Public Relations Inquiry, 8 (2), 109–125.

Bretas, V., & Alon, I. (2021). Franchising research on emerging markets: Bibliometric and content analyses. Journal of Business Research, 133 , 51–65.

Buhmann, A., & Fieseler, C. (2021). Towards a deliberative framework for responsible innovation in artificial intelligence. Technology in Society, 64 , 101475.

Carayannis, E. G., Christodoulou, K., Christodoulou, P., Chatzichristofis, S. A., & Zinonos, Z. (2022). Known unknowns in an era of technological and viral disruptions-implications for theory, policy, and practice. Journal of the Knowledge Economy, 13 (1), 587–610.

Cath, C. (2018). Governing artificial intelligence: Ethical, legal and technical opportunities and challenges. Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences, 376 , 20180080.

Cen, Z., & Wang, J. (2019). Crude oil price prediction model with long short term memory deep learning based on prior knowledge data transfer. Energy, 169 , 160–171.

Chattopadhyay, P. B., & Rangarajan, R. (2014). Application of ANN in sketching spatial nonlinearity of unconfined aquifer in agricultural basin. Agricultural Water Management, 133 , 81–91.

Chen, Z. S. (2022). Artificial Intelligence-virtual trainer: Innovative didactics aimed at personalized training needs. Journal of the Knowledge Economy , 19 . https://doi.org/10.1007/s13132-022-00985-0

Chen, H.-J., Huang, S.-Y., & Kuo, C.-L. (2009). Using the artificial neural network to predict fraud litigation: Some empirical evidence from emerging markets. Expert Systems with Applications, 36 (2, Part 1), 1478–1484.

Chou, J-S., & Pham, A-D. (2017). Nature-inspired metaheuristic optimization in least squares support vector regression for obtaining bridge scour information. Information Sciences, 399 , 64–80.

Cobo, M. J., López-Herrera, A. G., Herrera-Viedma, E., & Herrera, F. (2011). An approach for detecting, quantifying, and visualizing the evolution of a research field: A practical application to the Fuzzy Sets Theory field. Journal of Informetrics, 5 (1), 146–166.

Cockburn, I. M., Henderson, R., & Stern, S. J. N. C. (2018). The impact of artificial intelligence on innovation: An exploratory analysis. NBER Chapters.

Coglianese, C., & Lehr, D. (2017). Regulating by robot: Administrative decision making in the Machine-learning era. Georgetown Law Journal, 105 , 1147–1223.

Davenport, T. H., & Ronanki, R. (2018). Artificial intelligence for the real world. Harvard Business Review, 96 (1), 108–116.

de Saille, S. (2015). Innovating innovation policy: The emergence of ‘Responsible Research and Innovation.’ Journal of Responsible Innovation, 2 , 152–168.

Determann, L. (2018). No one owns data. SSRN Electronic Journal . https://doi.org/10.2139/ssrn.3123957

Article   Google Scholar  

Donthu, N., Kumar, S., Mukherjee, D., Pandey, N., & Lim, W. M. (2021). How to conduct a bibliometric analysis: An overview and guidelines. Journal of Business Research, 133 , 285–296.

Dopico, M., Gomez, A., Fuente, D., García, N., Rosillo, R., & Puche Regaliza, J. (2016). A Vision of Industry 4.0 from an Artificial Intelligence . Conference: The 2016 World Congress in Computer Science, Computer Engineering & Applied Computing (WORLDCOMP 2016) - International Conference on Artificial Intelligence (IC-AI 2016). At: Las Vegas, EEUU.

Dwivedi, Y. K., Hughes, D. L., Coombs, C., Constantiou, I., Duan, Y., Edwards, J. S., & Upadhyay, N. (2020). Impact of COVID-19 pandemic on information management research and practice: Transforming education, work and life. International Journal of Information Management, 55 , 102211.

Forliano, C., De Bernardi, P., & Yahiaoui, D. (2021). Entrepreneurial universities: A bibliometric analysis within the business and management domains. Technological Forecasting and Social Change, 165 , 120522.

Frank, M., Autor, D., Bessen, J., Brynjolfsson, E., Cebrian, M., Deming, D., & Rahwan, I. (2019). Toward understanding the impact of artificial intelligence on labor. Proceedings of the National Academy of Sciences of the United States of America, 116 , 6531–6539.

Frey, C. B., & Osborne, M. A. (2017). The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114 , 254–280.

Garfield, E. (1979). Is citation analysis a legitimate evaluation tool? Scientometrics, 1 (4), 359–375.

Gasteiger, E., & Prettner, K. (2017). On the possibility of automation-induced stagnation. Hohenheim Discussion Papers in Business, Economics and Social Sciences, University of Hohenheim, Faculty of Business, Economics and Social Sciences.

Gaur, A., & Kumar, M. (2018). A systematic approach to conducting review studies: An assessment of content analysis in 25years of IB research. Journal of World Business, 53 (2), 280–289.

Goertzel, B., Goertzel, T., & Goertzel, Z. (2017). The global brain and the emerging economy of abundance: Mutualism, open collaboration, exchange networks and the automated commons. Technological Forecasting and Social Change, 114 , 65–73.

Gomes, M. G., da Silva, V. H. C., Pinto, L. F. R., Centoamore, P., Digiesi, S., Facchini, F., & Neto, GCd. O. (2020). Economic, environmental and social gains of the implementation of artificial intelligence at dam operations toward Industry 4.0 principles. Sustainability, 12 (9), 3604.

Haefner, N., Wincent, J., Parida, V., & Gassmann, O. (2021). Artificial intelligence and innovation management: A review, framework, and research agenda. Technological Forecasting and Social Change, 162 , 120392.

He, S. W., Song, R., & Chaudhry, S. S. (2014). Service-oriented intelligent group decision support system: Application in transportation management. Information Systems Frontiers, 16 (5), 939–951.

Hernández-Callejo, L., Baladrón, C., Aguiar, J., Carro, B., Sanchez, A., & Lloret, J. (2013). Short-term load forecasting for microgrids based on artificial neural networks. Energies, 2013 . https://doi.org/10.3390/en6031385

Heylighen, F. (2017). Towards an intelligent network for matching offer and demand: From the sharing economy to the global brain. Technological Forecasting and Social Change, 114 , 74–85.

Hildebrandt, M. (2018). Law as computation in the era of artificial legal intelligence: Speaking law to the power of statistics. University of Toronto Law Journal, 68 , 12–35.

Hu, L., Miao, Y., Wu, G., Hassan, M., & Humar, I. (2018). iRobot-Factory: An intelligent robot factory based on cognitive manufacturing and edge computing. Future Generation Computer Systems, 90 . https://doi.org/10.1016/j.future.2018.08.006

Huang, M.-H., Rust, R., & Maksimovic, V. (2019). The feeling economy: Managing in the next generation of artificial intelligence (AI). California Management Review, 61 (4), 43–65.

Huin, S. F., Luong, L. H. S., & Abhary, K. (2003). Knowledge-based tool for planning of enterprise resources in ASEAN SMEs. Robotics and Computer-Integrated Manufacturing, 19 , 409–414.

Hutchinson, P. (2021). Reinventing innovation management: The impact of self-innovating artificial intelligence. IEEE Transactions on Engineering Management, 68 (2), 628–639.

Huynh, T. L. D., Hille, E., & Nasir, M. A. (2020). Diversification in the age of the 4th industrial revolution: The role of artificial intelligence, green bonds and cryptocurrencies. Technological Forecasting and Social Change, 159 , 120188.

Iandolo, F., Loia, F., Fulco, I., Nespoli, C., & Caputo, F. (2021). Combining big data and Artificial Intelligence for managing collective knowledge in unpredictable environment-insights from the Chinese case in facing COVID-19. Journal of the Knowledge Economy, 12 (4), 1982–1996.

Jakšič, M., & Marinc, M. (2019). Relationship banking and information technology: The role of artificial intelligence and fintech. Risk Management, 21 , 1–18.

Ju, K., Su, B., Zhou, D., & Zhang, Y. (2016). An incentive-oriented early warning system for predicting the co-movements between oil price shocks and macroeconomy. Applied Energy, 163 , 452–463.

Kak, A. (2018). The emergence of the personal data protection bill, 2018: A critique. Economic and Political Weekly, 53 , 12–16.

Kakatkar, C., Bilgram, V., & Füller, J. (2020). Innovation analytics: Leveraging artificial intelligence in the innovation process. Business Horizons, 63 (2), 171–181.

Kıran, M. S., Özceylan, E., Gündüz, M., & Paksoy, T. (2012). Swarm intelligence approaches to estimate electricity energy demand in Turkey. Knowledge-Based Systems, 36 , 93–103.

Lam-Gordillo, O., Baring, R., & Dittmann, S. (2020). Ecosystem functioning and functional approaches on marine macrobenthic fauna: A research synthesis towards a global consensus. Ecological Indicators, 115 , 106379.

Lee, C., & Lim, C. (2021). From technological development to social advance: A review of Industry 4.0 through machine learning. Technological Forecasting and Social Change, 167 , 120653.

Li, B., Hou, B., Yu, W., Lu, X., & Yang, C. (2017). Applications of artificial intelligence in intelligent manufacturing: A review. Frontiers of Information Technology & Electronic Engineering, 18 (1), 86–96.

Li, J., Wang, R., Wang, J., & Li, Y. (2018). Analysis and forecasting of the oil consumption in China based on combination models optimized by artificial intelligence algorithms. Energy, 144 , 243–264.

Liu, J., Chang, H., Forrest, J.Y.-L., & Yang, B. (2020). Influence of artificial intelligence on technological innovation: Evidence from the panel data of china’s manufacturing sectors. Technological Forecasting and Social Change, 158 , 120142.

Liu, L., Huang, J., & Yu, S. (2016). Prediction of primary energy demand in China based on AGAEDE optimal model. Chinese Journal of Population Resources and Environment, 14 (1), 16–29.

Liu, Y., Ma, X., Shu, L., Hancke, G. P., & Abu-Mahfouz, A. M. (2021). From Industry 4.0 to Agriculture 4.0: Current status, enabling technologies, and research challenges. IEEE Transactions on Industrial Informatics, 17 (6), 4322–4334.

Lu, H., Li, Y., Chen, M., Kim, H., & Serikawa, S. (2018). Brain intelligence: Go beyond artificial intelligence. Mobile Networks and Applications, 23 (2), 368–375.

Luo, Y., Xiao, Y., Cheng, L., Peng, G., & Yao, D. (2021). Deep learning-based anomaly detection in cyber-physical systems: Progress and opportunities. ACM Computing Surveys, 54 , 1–36.

Makridakis, S. (2017). The forthcoming Artificial Intelligence (AI) revolution: Its impact on society and firms. Futures, 90 , 46–60.

Mania, K. (2022). Legal technology: Assessment of the legal tech industry's potential. Journal of the Knowledge Economy , 25 . https://doi.org/10.1007/s13132-022-00924-z

McClure, P. K. (2017). “You’re fired”, says the robot: The rise of automation in the workplace, technophobes, and fears of unemployment. Social Science Computer Review, 36 (2), 139–156.

Mehmood, R., Alam, F., Albogami, N. N., Katib, I., Albeshri, A., & Altowaijri, S. M. (2017). UTiLearn: A personalised ubiquitous teaching and learning system for smart societies. IEEE Access, 5 , 2615–2635.

Mhlanga, D. (2020). Industry 4.0 in finance: The impact of artificial intelligence (AI) on digital financial inclusion. International Journal of Financial Studies, 8 (3), 45.

Moll, J., & Yigitbasioglu, O. (2019). The role of internet-related technologies in shaping the work of accountants: New directions for accounting research. The British Accounting Review, 51 (6), 100833.

Naimi-Sadigh, A., Asgari, T., & Rabiei, M. (2021). Digital transformation in the value chain disruption of banking services. Journal of the Knowledge Economy , 31 . https://doi.org/10.1007/s13132-021-00759-0

Nasr, M. M., Anwar, S., Al-Samhan, A. M., Ghaleb, M., & Dabwan, A. (2020). Milling of graphene reinforced Ti6Al4V nanocomposites: An artificial intelligence based Industry 4.0 approach. Materials, 13 (24), 5707.

Nemitz, P. (2018). Constitutional democracy and technology in the age of artificial intelligence. Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences, 376 (2133), 20180089.

Niu, Y. F. (2018). The application of artificial intelligence and intelligent decision in men volleyball’s lineup tactics. Lecture Notes in Real-Time Intelligent Systems, 613 , 263–270.

Ozturk, T., Talo, M., Yildirim, E. A., Baloglu, U. B., Yildirim, O., & Rajendra Acharya, U. (2020). Automated detection of COVID-19 cases using deep neural networks with X-ray images. Computers in Biology and Medicine, 121 , 103792.

Peres, R. S., Jia, X., Lee, J., Sun, K., Colombo, A. W., & Barata, J. (2020). Industrial artificial intelligence in Industry 4.0 - Systematic review, challenges and outlook. IEEE Access, 8 , 220121–220139.

Pham-Duc, B., Tran, T., Le, H.-T.-T., Nguyen, N.-T., Cao, H.-T., & Nguyen, T.-T. (2021) Research on Industry 4.0 and on key related technologies in Vietnam: A bibliometric analysis using Scopus. Learned Publishing. https://doi.org/10.1002/leap.1381

Pinter, J., Fels, M., Lycon, D. S., Meeuwig, J. W., & Meeuwig, D. J. (1995). An intelligent decision support system for assisting industrial wastewater management. Annals of Operations Research, 58 , 455–477.

Qin, Y., Wang, X. X., Xu, Z. S., & Škare, M. (2021). The impact of poverty cycles on economic research: Evidence from econometric analysis. Economic Research-Ekonomska Istraživanja, 34 (1), 152–171.

Qin, Y., Xu, Z. S., Wang, X. X., & Škare, M. (2020). Are family firms in the eyes of economic policy? International Entrepreneurship and Management Journal . https://doi.org/10.1007/s11365-020-00699-2

Qin, Y., Xu, Z. S., Wang, X. X., & Škare, M. (2022). Green energy adoption and its determinants: A bibliometric analysis. Renewable and Sustainable Energy Reviews, 153 , 111780.

Rodriguez, G. G., Gonzalez-Cava, J. M., & Perez, J. A. M. (2020). An intelligent decision support system for production planning based on machine learning. Journal of Intelligent Manufacturing, 31 (5), 1257–1273.

Rust, R. T. (2020). The future of marketing. International Journal of Research in Marketing, 37 (1), 15–26.

Ryman-Tubb, N. F., Krause, P., & Garn, W. (2018). How Artificial Intelligence and machine learning research impacts payment card fraud detection: A survey and industry benchmark. Engineering Applications of Artificial Intelligence, 76 , 130–157.

Samid, G. (2021). Artificial intelligence assisted innovation. https://doi.org/10.5772/intechopen.96112

Sanz, E., Blesa, J., & Puig, V. (2021). BiDrac Industry 4.0 framework: Application to an Automotive Paint Shop Process. Control Engineering Practice, 109 , 104757.

Shakir, M., Mehmood, F., Bibi, Z., & Anjum, M. (2019). Innovation and artificial intelligence. https://doi.org/10.1007/978-3-030-20454-9_31

Sharabov, M., & Tsochev, G. (2020). The use of artificial intelligence in Industry 4.0. Problems of Engineering Cybernetics and Robotics, 73 . https://doi.org/10.7546/PECR.73.20.02

Skiba, M., Mrówczyńska, M., & Bazan-Krzywoszańska, A. (2017). Modeling the economic dependence between town development policy and increasing energy effectiveness with neural networks. Case study: The town of Zielona Góra. Applied Energy, 188 , 356–366.

Skrop, A. (2018). Industry 4.0 - Challenges in industrial artificial intelligence. Conference: II. International Scientific Conference on Tourism and Security. At: Hungary.

Soh, Y. W., Koo, C. H., Huang, Y. F., & Fung, K. F. (2018). Application of artificial intelligence models for the prediction of standardized precipitation evapotranspiration index (SPEI) at Langat River Basin, Malaysia. Computers and Electronics in Agriculture, 144 , 164–173.

Su, H.-N., & Lee, P.-C. (2010). Mapping knowledge structure by keyword co-occurrence: A first look at journal papers in Technology Foresight. Scientometrics, 85 (1), 65–79.

Syam, N., & Sharma, A. (2018). Waiting for a sales renaissance in the fourth industrial revolution: Machine learning and artificial intelligence in sales research and practice. Industrial Marketing Management, 69 , 135–146.

Tang, X., Li, X., Ding, Y., Song, M., & Bu, Y. (2020). The pace of artificial intelligence innovations: Speed, talent, and trial-and-error. Journal of Informetrics, 14 (4), 101094.

Tariq, S., Hu, Z., & Zayed, T. (2021). Micro-electromechanical systems-based technologies for leak detection and localization in water supply networks: A bibliometric and systematic review. Journal of Cleaner Production, 289 , 125751.

Trifan, B., & Buzatu, A. (2020). Sustainable business enhanced through digital transformation and artificial intelligence in the context of Industry 4.0. Conference: BASIQ. At: Italy.

Tsiknakis, N., Trivizakis, E., Vassalou, E., Papadakis, G., Spandidos, D., Tsatsakis, A., & Marias, K. (2020). Interpretable artificial intelligence framework for COVID‑19 screening on chest X‑rays. Experimental and Therapeutic Medicine, 20 . https://doi.org/10.3892/etm.2020.8797

Turner Lee, N. (2018). Detecting racial bias in algorithms and machine learning. Journal of Information, Communication and Ethics in Society, 16 (3), 252–260.

Uzlu, E., Kankal, M., Akpınar, A., & Dede, T. (2014). Estimates of energy consumption in Turkey using neural networks with the teaching–learning-based optimization algorithm. Energy, 75 , 295–303.

Vallaster, C., Kraus, S., MerigóLindahl, J. M., & Nielsen, A. (2019). Ethics and entrepreneurship: A bibliometric study and literature review. Journal of Business Research, 99 , 226–237.

von Joerg, G., & Carlos, J. (2022). Design framework for the implementation of AI-based (service) business models for small and medium-sized manufacturing enterprises. Journal of the Knowledge Economy , 19 . https://doi.org/10.1007/s13132-022-01003-z

van Eck, N. J., & Waltman, L. (2010). Software survey: VOSviewer, a computer program for bibliometric mapping. Scientometrics, 84 (2), 523–538.

van Oorschot, J. A. W. H., Hofman, E., & Halman, J. I. M. (2018). A bibliometric review of the innovation adoption literature. Technological Forecasting and Social Change, 134 , 1–21.

Verganti, R., Vendraminelli, L., & Iansiti, M. (2020). Innovation and design in the age of artificial intelligence. Journal of Product Innovation Management, 37 (3), 212–227.

Vermeulen, B., Kesselhut, J., Pyka, A., & Saviotti, P. P. (2018). The impact of automation on employment: Just the usual structural Change? Sustainability, 10 (5), 1661.

Vivarelli, M. (2014). Innovation, employment and skills in advanced and developing countries: A survey of economic literature. Journal of Economic Issues, 48 (1), 123–154.

Vogt, J. (2021). Where is the human got to go? Artificial intelligence, machine learning, big data, digitalisation, and human-robot interaction in Industry 4.0 and 5.0. AI & Society . https://doi.org/10.1007/s00146-020-01123-7

Vyshnevskyi, O., Liashenko, V., & Amosha, O. (2019). The impact of Industry 4.0 and AI on economic growth. Scientific Papers of Silesian University of Technology Organization and Management Series, 9 , 391–400.

Wang, C., Lim, M. K., Zhao, L., Tseng, M.-L., Chien, C.-F., & Lev, B. (2020a). The evolution of Omega-The International Journal of Management Science over the past 40 years: A bibliometric overview. Omega, 93 , 102098.

Wang, L., Luo, G.-L., Sari, A., & Shao, X.-F. (2020b). What nurtures fourth industrial revolution? An investigation of economic and social determinants of technological innovation in advanced economies. Technological Forecasting and Social Change, 161 , 120305.

Wang, L., Zhang, H.-C., & Wang, Q. (2019). On the concepts of artificial intelligence and innovative design in product design. IOP Conference Series: Materials Science and Engineering, 573 , 012095.

Wang, X. X., Chang, Y. R., Xu, Z. S., Wang, Z. D., & Kadirkamanathan, V. (2021a). 50 Years of international journal of systems science: A review of the past and trends for the future. International Journal of Systems Science, 52 (8), 1515–1538.

Wang, X. X., Xu, Z. S., & Škare, M. (2020c). A bibliometric analysis of Economic Research-Ekonomska Istraživanja (2007–2019). Economic Research-Ekonomska Istraživanja, 33 (1), 865–886.

Wang, X. X., Xu, Z. S., Su, S.-F., & Zhou, W. (2021b). A comprehensive bibliometric analysis of uncertain group decision making from 1980 to 2019. Information Sciences, 547 , 328–353.

Weng, B., Martinez, W., Tsai, Y.-T., Li, C., Lu, L., Barth, J. R., & Megahed, F. M. (2018). Macroeconomic indicators alone can predict the monthly closing price of major U.S. indices: Insights from artificial intelligence, time-series analysis and hybrid models. Applied Soft Computing, 71 , 685–697.

Williams, P. (2019). Does competency-based education with blockchain signal a new mission for universities? Journal of Higher Education Policy and Management, 41 (1), 104–117.

Wolff, J. G. (2014). Big data and the SP theory of intelligence. IEEE Access, 2 , 301–315.

Xue, L., Zhu, Y. P., & Xue, Y. (2013). RAEDSS: An integrated decision support system for regional agricultural economy in China. Mathematical and Computer Modelling, 58 (3–4), 480–488.

Yamashiro, S. (1986). Online secure-economy preventive control of power systems by pattern recognition. IEEE Transactions on Power Systems, 1 (3), 214–219.

Yong, B., Xu, Z. J., Wang, X., Cheng, L. B., Li, X., Wu, X., & Zhou, Q. G. (2018). IoT-based intelligent fitness system. Journal of Parallel and Distributed Computing, 118 , 14–21.

Zhang, Y., Zhang, M., Li, J., Liu, G., Yang, M. M., & Liu, S. (2021). A bibliometric review of a decade of research: Big data in business research – Setting a research agenda. Journal of Business Research, 131 , 374–390.

Zheng, X., Le, Y., Chan, A. P. C., Hu, Y., & Li, Y. (2016). Review of the application of social network analysis (SNA) in construction project management research. International Journal of Project Management, 34 (7), 1214–1225.

Download references

The work was supported by the National Natural Science Foundation of China (No. 72071135), the Fundamental Research Funds for the Central Universities (YJ202063, SXYPY202146), and China Postdoctoral Science Foundation (No. 2021M692259).

Author information

Authors and affiliations.

Business School, Sichuan University, Chengdu, China

Yong Qin, Zeshui Xu & Xinxin Wang

Juraj Dobrila University of Pula, Pula, Croatia

Marinko Skare

You can also search for this author in PubMed   Google Scholar

Contributions

Yong Qin, Zeshui Xu, Xinxin Wang, and Marinko Škare conceived the study and were responsible for the data collection, design, and development of the data analysis as well as for data interpretation. Yong Qin, Zeshui Xu, and Xinxin Wang wrote the first draft of the article.

Corresponding author

Correspondence to Marinko Skare .

Ethics declarations

Conflict of interest.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This article is part of the Topical Collection on AI in the Knowledge Economy and Society: Implications for Theory, Policy and Practice

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Qin, Y., Xu, Z., Wang, X. et al. Artificial Intelligence and Economic Development: An Evolutionary Investigation and Systematic Review. J Knowl Econ 15 , 1736–1770 (2024). https://doi.org/10.1007/s13132-023-01183-2

Download citation

Received : 24 June 2021

Accepted : 21 February 2023

Published : 11 March 2023

Issue Date : March 2024

DOI : https://doi.org/10.1007/s13132-023-01183-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Artificial intelligence
• Economic development
• Bibliometrix
• Content analysis
• Bibliographic coupling

Advertisement

• Find a journal
• Publish with us
• Track your research

The Official Journal of the Pan-Pacific Association of Input-Output Studies (PAPAIOS)

• Open access
• Published: 09 September 2023

Implications of AI innovation on economic growth: a panel data study

• Julius Tan Gonzales   ORCID: orcid.org/0009-0002-4008-5961 1  

Journal of Economic Structures volume  12 , Article number:  13 ( 2023 ) Cite this article

17k Accesses

11 Citations

3 Altmetric

Metrics details

The application of artificial intelligence (AI) across firms and industries warrants a line of research focused on determining its overall effect on economic variables. As a general-purpose technology (GPT), for example, AI helps in the production, marketing, and customer acquisition of firms, increasing their productivity and consumer reach. Aside from these, other effects of AI include enhanced quality of services, improved work accuracy and efficiency, and increased customer satisfaction. Hence, this study aims to gauge the impact of AI on the economy, specifically on long-run economic growth. This study conjectures a positive relationship between AI and economic growth. To test this hypothesis, this study makes use of a panel dataset of countries from 1970 to 2019, and the number of AI patents as a measure of AI. A text search query is performed to distinguish AI patents from other types of innovations in a public database. Employing fixed effects and generalized method of moments (GMM) estimation, this paper finds a positive relationship between AI and economic growth, which is higher than the effect of the total population of patents on growth. Furthermore, other results indicate that AI’s influence on growth is more robust among advanced economies, and more evident towards the latter periods of the dataset.

1 Introduction

The developments in computer science and digital technology, including artificial intelligence (AI) and machine learning, naturally led to their application in key sectors such as healthcare, finance, manufacturing, and transport. Footnote 1 Their increasing use in industries has opened up questions as to whether these technologies may have an impact on economic variables. In neoclassical and endogenous economic growth models, for example, technical change brings about increases in productivity, leading to economic growth. Hence, breakthroughs in computing technology should also entail increases in growth rates.

In particular, the last 60 years have witnessed a shift in production from traditional inputs to more information and communications technology (ICT)-based, capital-intensive tools. The introduction of modern computers and the Internet in the early 1990s, and more recently, AI, led to changes in the methods of production. In keeping with the rise of new technologies, Zeira ( 1998 ) proposed an economic growth model adopting technological innovations that reduce labor inputs but require more capital. Footnote 2 Furthermore, more recent empirical studies on the subject have found these technologies as potential sources of economies of scale (e.g., Nightingale 2000 ; Wang et al. 2011 ; Nchake and Shuaibu 2022 ). Thus, it is only expected that advancements in ICT may have a positive effect on overall productivity and economic growth.

This study continues this line of research by exploring the relationship between technical change, as manifested by modern developments in science and technology, and economic growth. Several economic papers have been published on the subject, of proxying technology with various forms of knowledge and ICT measures (e.g., patents and scientific journals, Internet penetration, computer ownership, etc.). This paper is similar to past academic papers, yet with a special focus on AI as a newer form of technology.

Estimation results reveal a positive and significant impact of AI on long-run economic growth in a cross-country panel dataset. The magnitude of AI's effect on growth is also higher than that of total patents. Furthermore, the contribution of AI to growth is more robust both for advanced economies and for the latter half of the period considered in the estimation.

2 Literature review

Endogenous growth models are central to much of the existing literature on technology and economic growth. Arrow ( 1962 ) associates technology as a “by-product of ordinary production” through knowledge accumulation, aptly termed “learning-by-doing.” The process of learning through repetition and experience should manifest itself in increases in productivity, thus creating opportunities for economic growth. Footnote 3 Arrow ( 1962 ) assumes that technical change, born out of knowledge and experience, is embodied in new physical capital, which then enters the production process and improves “productive efficiency.” Footnote 4

The literature on technology and growth follows this line of thought, assuming that technological change increases capital productivity. In line with the Schumpeterian tradition, Zeira ( 1998 ) presents a theoretical framework involving intermediate goods in production. Technology adoption increases the intensity of capital and replaces labor in the production process. Technology is then adopted if it increases output; however, as technology requires more capital input, not all countries can keep up with the technological frontier. Footnote 5 The disparities in the levels of technology across countries then result in differences in overall output and productivity.

Meanwhile, Acemoglu and Restrepo ( 2018 ) have constructed a “task-based” framework, treating automation and the creation of new tasks as types of technological innovation. Both types of technology are necessary to increase productivity. Initially, Acemoglu and Restrepo ( 2018 ) considered that all tasks could be done by labor, whereas “lower-indexed” tasks could and would be automated. Footnote 6 However, automation requires some capital investment, thus raising the share of capital and decreasing the share of labor in production. This is counterbalanced, though, by creating new and more sophisticated tasks, where labor has a “comparative advantage.” In the long run, there is a “stable, balanced growth path” where the two types of innovations coexist and grow at the same rate. Footnote 7

The aforementioned theoretical works help explain the relationship between modern science and ICT developments and economic growth. However, empirical evidence on more recent forms of technological innovation, such as AI and machine learning, is still limited. This can be attributed to an insufficient amount of data both at the firm and at macro levels, especially when dealing with long-run growth. Footnote 8 This study is an attempt to contribute to the body of literature on this subject despite limitations on data availability. In the following discussion in this section, this paper will revisit some recent publications regarding the relationship between technical change and economic growth, using common scientific knowledge and technology variables.

In empirical studies, the number of patents and scientific journals are common measures of technological innovation. In the Schumpeterian context, patents represent ownership of monopoly rents from the invention of new technology. Firms aim for exclusive rights over these monopoly rents; thus, new technologies that improve productivity are continuously invented, while dismantling obsolete ones in the process—the so-called “creative destruction.” As better technologies are created, firms become more productive, possibly achieving increasing returns to scale status. Footnote 9 Hence, countries with higher concentrations of patents may signal higher productivity and levels of production, and of course, national growth. Footnote 10

On the other hand, scientific journals index the level of research and development (R&D). Based on standard growth models, technical progress is a product of knowledge accumulation, which is made possible through continuous R&D efforts. According to Kim and Lee ( 2015 ), academic articles as a measure of scientific knowledge have been regarded as a contributor to economic growth, citing scientific journals published by institutions and universities as sources of “patents and industrial technology.” Assuming academic knowledge from journal articles can be transformed into concrete technological inputs for production, published research should then also contribute to overall productivity and growth. Footnote 11

However, Kim and Lee ( 2015 ) conclude that it is patents and not scientific journals that contribute to economic growth. They considered academic articles to be sources of scientific knowledge, whereas patents are embodiments of technological knowledge. Technological knowledge, though, is more a product of the private R&D efforts of firms than of scientific research from academic institutions. Using panel data estimation and evidence from Latin American economies, Kim and Lee ( 2015 ) found an insignificant effect of scientific knowledge, while patents indicated significant and positive impacts on economic growth.

Studies about patents and growth are numerous, often arriving at similar results (e.g., Lach 1995 ; Sinha 2008 ; Kim et al. 2012 ). In an earlier work modeling innovation and entrepreneurship with economic growth, Wong et al. ( 2005 ) found a significant and positive effect of patent grants as an indicator of innovation on country growth rates. In contrast, recent studies such as those by Sweet and Eterovic ( 2019 ) and Blind et al. ( 2022 ) found no significant effect of patents on economic growth.

In another recent study, Nguyen and Doytch ( 2022 ) found a positive and significant effect of total patents on economic growth for advanced economies, but the magnitude of the effect of the technology variable weakens for emerging economies. Footnote 12 Moreover, ICT patents only contribute to economic growth among advanced economies. In addition, the authors found that total patents, regardless of domain, are not significant in the long run, but ICT patents remain positive and significant.

On the other hand, studies on the effect of scientific research, measured by the number of scientific journals, on cross-country growth tend to be mixed. As mentioned previously, Kim and Lee ( 2015 ) discovered no significant impact of scientific knowledge from academic articles on growth. Meanwhile, Ntuli et al. ( 2015 ) found differing results in determining causality between research output and growth among OECD countries. Research output exhibits “unidirectional causality” on growth in some countries such, as the United States, Finland, Hungary, and Mexico, but is negligible in other OECD members.

Existing literature suggests a weak or ambiguous relationship between academic research and national growth (e.g., Inglesi-Lotz et al. 2014 ; Hatemi-J et al. 2016 ). Footnote 13 In contrast, Solarin and Yen ( 2016 ) obtained a positive relationship between research publications and economic growth using a cross-country panel dataset. They found that the effect was significant “irrespective of whether the focus is on developed countries or developing nations.” However, Solarin and Yen ( 2016 ) noted that the impact on growth is stronger in advanced economies.

Interestingly, Mueller ( 2006 ) found that research output may be favorable to local economic performance. Mueller ( 2006 ) analyzed the impact of private industry and university R&D, along with measures of entrepreneurship and university-industry relations, Footnote 14 on regional aggregate output in West Germany. Regression results have established individual, positive effects of each variable on regional economic performance.

Further, at the firm level, “intangible assets” such as “R&D, goodwill, brand equity, patents, copyrights, software, licenses, image, and organization” are “enhancers” of total factor productivity (TFP) (Nakatani 2021 ). Comparing firms within the ICT sector across five countries, intangible assets revealed a significant impact, though differing in magnitude, on the productivity of ICT firms across countries. Footnote 15 This can be attributed to some countries already being at the forefront of the global technology frontier. Hence, the additional effect of intangible assets on firm productivity diminishes (Nakatani 2021 ).

This study, however, is more interested in a specific technological innovation different from ICT, namely AI, including machine learning. With AI swiftly becoming the new general-purpose technology (GPT) (Trajtenberg 2018 ), comparisons between AI and previous technologies, particularly ICT, have been raised (Lu and Zhou 2021 ). However, AI is considered to “impact a broader range of sectors,” leading to “different implications at the aggregate level” and an “unpredictable future development.” Furthermore, ICT is known to require high investments in capital over long periods, whereas AI can leverage data and cloud services that can help lower capital investments. These differences could potentially lead to a distinct “pathway” for AI adoption, different from that of previous technologies (Lu and Zhou 2021 ).

Because of the scarcity of data, there is a dearth of empirical evidence on the topic of AI as a driver of economic growth. Nonetheless, this article attempts to determine this relationship using an available measure that can indicate the level of AI per country.

2.1 What is AI?

AI encompasses a broad category of technology, and there is not a single, widely accepted definition. However, international organizations have similar definitions of AI. The European Parliamentary Research Service (EPRS), for example, refers to AI as machines that perform “human-like cognitive processes,” namely, “learning, understanding, reasoning and interacting.” As a general-purpose technology, AI can take many forms such as a “technical infrastructure (i.e., algorithms), a part of the (production) process, or an end-user product” (Szczepański 2019 ). Hence, in contrast with traditional technologies that automate routine processes, AI technologies even go further to mimic human activities that require cognition, and their application and use are not limited to the production process.

Meanwhile, the International Telecommunication Union (ITU) broadly defines AI as “self-learning, adaptive systems.” Accordingly, there are several “approaches” in defining AI, namely: (1) in terms of “technologies, techniques and/or approaches” such as “a neural network approach to machine translation”; (2) in terms of “purpose,” which include facial and image recognition; (3) in terms of “functions,” such as the “ability to understand language, recognize pictures, solve problems, and learn”; and (4) in terms of “agents or machines or algorithms” such as robots and self-driving cars (International Telecommunication Union 2023 ).

Furthermore, Montagnier and Ek ( 2021 ) cite several definitions of AI by individual countries and organizations such as the European Commission and the OECD. For instance, the OECD defines AI as a “machine-based system” that can “make predictions, recommendations, or decisions” and “operate with varying levels of autonomy” (Yeung 2020 ). Additionally, the European commission ( 2021 ) provides some examples of AI, which include “chatbots” and “virtual assistants,” “face recognition systems,” “machine translation software,” “data analysis based on machine learning,” “autonomous robots,” and “autonomous drones.” On the other hand, national statistics institutions such as the French Institut national de la statistique et des études économiques (INSEE) ( 2019 ) describe AI as “technologies” that can perform “cognitive tasks traditionally performed by humans,” whereas Statistics Sweden ( 2020 ) notes that physically, AI may be “purely software based or embedded in hardware.”

Because of its broad definition and the lack of a single, universally accepted descriptor of AI, classifying existing AI technologies is also a difficult task. In spite of this, Sarker ( 2022 ) categorized AI into five types, which include analytical, functional, interactive, textual, and visual. Footnote 16 However, the most commonly heard terms in AI are the “techniques” used in developing intelligent and smart systems in various real-world application areas.” Sarker ( 2022 ) identified at least ten “potential categories,” namely:

Machine learning,

Neural network and deep learning (including generative AI),

Data mining, knowledge discovery, and advanced analytics,

Rule-based modeling and decision-making,

Fuzzy logic-based approach,

Knowledge representation, uncertainty reasoning, and expert system modeling,

Case-based reasoning,

Text mining and natural language processing,

Visual analytics, computer vision, and pattern recognition,

Hybrid approach, searching, and optimization.

While each AI technique has its scope and specific applications, it is often that existing technologies are combinations and applications of various categories. Thus, grouping AI systems according to specific types or techniques is not always feasible. Moreover, AI development is a wide and ongoing practice, and more and newer forms of AI technologies are continuously produced over time. For example, ChatGPT, a form of generative AI technology that employs deep learning, was released to the public in 2022, and quickly became a groundbreaking AI technology due to its ability to interact with individuals and provide “comprehensive and practical responses” (Marr 2023 ). Footnote 17 ChatGPT is built upon “foundational large language models” (LLMs), which go beyond conventional natural language algorithms.

In addition, AI development may be unique to its industry due to the nature of AI itself. Coiera ( 2019 ) identifies three main stages of AI development, termed “miles.” The “first mile” consists of data acquisition, pre-processing, or “cleaning.” The “middle mile” includes “developing and testing the technical performance of different algorithms” that are built using the data acquired in the first stage. After all tests and tuning are completed, an AI system enters the last mile, where it is “embedded in real-world processes and tested for impact on real-world outcomes.”

However, each stage of AI development has its challenges. The first mile entails “foundational challenges,” such as “gathering and curating” huge amounts of high-quality data. Acquiring large amounts of data presents a potential “bottleneck,” and “translates into a roadblock to technology application.” Meanwhile, the middle mile involves the difficulties of “data-driven algorithm development,” such as “biases, replicability, causal inference, avoiding overfitting on training data, and enhancing the generalizability of any models and algorithms” (Coiera 2019 ). Footnote 18

Finally, and likely the hardest task, occurs in the third mile. As it turns out, “AI does not do anything on its own”; therefore, AI systems must somehow “connect” to the real world. Simply, the impact of an AI system must be “consequential” and “meaningful.” For example, the current setting does not necessitate better diagnoses of cancer but “more nuanced” and “less aggressive” approaches to detection and management. Hence, the last mile refers to the implementation of AI itself in real-world processes. AI implementation faces a plethora of challenges, which can be classified under “measurement,” “generalization and calibration,” and “local context” (Coiera 2019 ). Footnote 19

2.2 AI and economic growth

AI drives economic growth by stimulating gains both from the supply side and the demand side. AI can drive business productivity through (1) automation of processes with the use of robots and “autonomous vehicles,” and (2) improvements in the existing labor force by equipping them with AI technologies. On the other hand, AI can generate an increase in consumer demand with the availability of “personalised and/or higher-quality” products and services. Accordingly, it is expected that AI could contribute up to USD 15.7 trillion to the global economy in 2030 (Rao and Verweij 2017 ).

Furthermore, the contributions of AI may be specific to the sectors where it is applied, such as manufacturing, health, finance, energy, and transport. For example, AI supports healthcare services through early detection and diagnosis of illnesses, identification of “potential pandemics and tracking incidence,” and “imaging diagnostics” in radiology and pathology. Meanwhile, AI contributions to the financial sector include applications for fraud detection and anti-money laundering. Also, AI developments such as “robo-advice” make “customized investment solutions” possible in managing financial goals and optimizing clients’ funds. In addition, AI enables “autonomous trucking and delivery,” traffic control systems, and improved security in the transport sector (Rao and Verweij 2017 ).

Recently, Lu ( 2021 ) built a theoretical framework that traces the impact of AI on endogenous growth. Lu ( 2021 ) likens AI to human capital accumulation, “as it can learn and accumulate knowledge by itself.” Secondly, AI is a “nonrival input,” which can be used in production without having it “detract from its ability to accumulate AI.” This implies that AI is disembodied from physical capital, and should be considered a separate input. Footnote 20 Moreover, Lu ( 2021 ) unveils a balanced growth path in the three-sector endogenous growth model, where production and factors including AI grow at the same rate. Footnote 21

Using provincial data from China, He ( 2019 ) estimated the effect of AI on regional economic growth. Unlike most innovation studies on ICT and growth, He ( 2019 ) makes use of fixed assets investment in ICT to GDP as a measure of AI, Footnote 22 rather than AI-specific patents or published articles. Similarly, Fan and Liu ( 2021 ) tested AI as a tool for the sustainable economic development of Chinese provinces. Footnote 23 The results in both studies are consistent with theories on the growth-enhancing capability of AI.

Furthermore, Yang ( 2022 ) evaluated the effect of both AI and non-AI patents on firm-level productivity and employment in Taiwan. Both types of patents were found to improve productivity and employment among Taiwanese electronic firms. Estimation results revealed that both AI and non-AI patents contribute to TFP, and the difference in elasticities between the two patent types is insignificant. Moreover, when TFP is replaced by labor productivity, the estimated coefficient for AI patents is lower than in the model with TFP as a dependent variable. Yang ( 2022 ) suggested that this can be attributed to AI technology having a “greater effect on capital productivity,” which is consistent with the frameworks of Arrow ( 1962 ) and Zeira ( 1998 ).

At present, there are limited empirical works regarding AI as an engine of economic growth, primarily because of the unavailability of data. Footnote 24 Though extant literature on the topic finds a positive relationship between AI technology and economic growth, general sentiment suggests the effect of AI on growth is complex (He 2019 ) and difficult to measure. Intuitively, this can be because of its multifaceted role as an input to production. Still, with the increasing use of AI across countries and industries, this article seeks to measure the impact of AI on national growth rates amidst empirical constraints.

3 Theoretical framework

This study follows an endogenous growth framework. An endogenous model of economic growth often starts with the basic Cobb–Douglas function. However, this study also takes into account human capital as an input to production:

where \(Y\) is the total output, \(K\) stands for capital, \(L\) for labor, and \(H\) is human capital. The elasticities of output to capital, labor, and human capital are denoted by \(\alpha\) , \(\beta\) , and \(\gamma\) , respectively. Meanwhile, \(A\) is the level of knowledge, or as proposed by Jones and Williams ( 1998 ), the stock of ideas, available in an economy.

To obtain the output per unit of labor, Eq. ( 1 ) is divided on both sides by \(L\) . Multiplying the right-hand side with \(\frac{{L}^{\alpha +\gamma }}{{L}^{\alpha +\gamma }}=\frac{{L}^{\alpha }}{{L}^{\alpha }}\cdot \frac{{L}^{\gamma }}{{L}^{\gamma }}=1\) and assuming constant returns to scale, \(\alpha +\beta +\gamma =1\) , results in Eq. ( 2 ):

For simplicity, the per unit of labor variables are replaced by small letters, as with Eq. ( 3 ):

The technology factor \(A\) is seen as the available knowledge stock at time \(t\) . Romer ( 1990 ) proposed that since knowledge is a nonrival input, all researchers can utilize existing knowledge stock at the same time. Summing across all individual efforts in research yields Eq. ( 4 ):

where \(R\) is the research effort or resources devoted to research. The function is assumed to be increasing in \(R\) , as more research leads to more ideas. Jones and Williams ( 1998 ), though, noted that Eq. ( 4 ) may be increasing or decreasing in \(A\) , depending on how previous ideas affect current research.

A basic (and crucial) assumption is that the parameter \(\theta\) is assumed to be 1, to show that the increase in \(R\) results in an increase in new ideas. Footnote 25 Meanwhile, the coefficient \(\delta\) depicts the productivity of research, as proposed by Romer ( 1990 ) and Jones and Williams ( 1998 ).

To estimate Eq. ( 3 ), the equation is transformed into its natural log form. Further, the differenced natural logged form of Eq. ( 3 ) is obtained to calculate the growth rate:

The growth rate of \(y\) is defined as \({g}_{y}=\frac{\dot{y}}{y}\) , where \(\dot{y}=\frac{dy}{dt}\) . The term \(\dot{y}\) represents the difference, or change, in output per worker between two time periods (the change in \(t\) ). Mathematically, the growth rate can be further expressed as \({g}_{y}=\frac{dy/dt}{y}=\frac{d\mathrm{ln}y}{dt}= \frac{\mathrm{ln}{y}_{t}-{\mathrm{ln}y}_{t-s}}{s}\) . Therefore, dividing Eq. ( 5 ) by the change in \(t\) yields the growth rate equation: Footnote 26

Substituting Eq. ( 4 ) for the value of \(\dot{A}\) in Eq. ( 6 ) and simplifying the resulting equation yields:

Finally, the growth rate of \(y\) can be written as:

This study focuses on determining the relationship between AI innovation and economic growth. Thus, the variable \(R\) is proxied by the level of AI innovation in the economy, given by the amount of AI patents published within a certain period. Notably, this is slightly different from the theoretical specification, which indicates \(R\) as inputs or resources devoted to research (e.g., R&D expenditure, share of labor assigned to R&D, etc.). In general, patents are precisely the output of these R&D efforts. The choice of R&D input, such as the number of researchers, or output, such as the number of patents, in economic analysis, has been discussed by Griliches ( 1998 ). Ultimately, this decision depends on the size of the error terms in the relationships among patents, research, and knowledge stock. Footnote 27 Moreover, Griliches ( 1998 ) conjectures that if the “stochastic component” of knowledge stock is captured to some extent by patenting, using patents may have some “value added” over the use of common research inputs as an indicator of knowledge.

Patents embody the quantity, type, inventiveness, and complexity of innovation created in a given time (Griliches 1998 ). Although not without disadvantages, patents can serve as a good indicator of technical knowledge. More importantly, patent data are more readily available for analysis than research input measures, especially for AI. Hence, this study makes use of the number of AI and total patents as a proxy for R&D.

Furthermore, while the model discussed in this section explains how traditional research translates to economic growth, the current model might not fully encapsulate the effect of AI. Footnote 28 As stated previously, the employed model assumes constant returns to research (and by extension, AI). However, because of the nonrivalry of data and the possibility of AI “outpacing” human intelligence, continuous AI invention may exhibit increasing returns, further leading to a “technological singularity,” or explosion of growth rates (Aghion et al. 2018 ). Exploring empirical evidence of such a mechanism is beyond the scope of this study; however, it is a highly recommended topic for future research. Footnote 29

4 Data and methodology

For this study, the primary challenge to perform econometric analysis is obtaining data that can measure the level of AI in a cross-country, panel dataset format. As discussed in the previous sections, the most common indicator of technological innovation is patent publications. Therefore, this study uses AI patents as a measure of AI.

Data for AI patents are available from the Google Patents Public Data, provided by the Information for Industry, Inc. (IFI) CLAIMS Patent Services. To identify AI patents, a text search query was performed in the patents database. The text search includes common words or phrases related to AI, such as “artificial intelligence,” “face recognition,” “virtual assistant,” “machine learning,” etc. Footnote 30 Meanwhile, data for the dependent and control variables are sourced from the United Nations (UN) Department of Economic and Social Affairs Statistics Division and the World Bank.

This study echoes the econometric models of Wong et al. ( 2005 ), Kim and Lee ( 2015 ), and He ( 2019 ) among others. Estimating Eq. ( 8 ) from the previous section, the econometric model follows the equation:

where \({\text{Growth}}_{it}\) is the annual average real GDP per capita growth rate of country \(i\) over a certain period \(t\) , i.e., five years, calculated by dividing the difference between the natural log value of end-of-period real GDP per capita (in USD and constant 2015 prices) and the natural log value of initial real GDP, by the number of years in period \(t\) . Hence, \({\text{Growth}}_{it}\) is the instantaneous growth rate of the real GDP of country \(i\) in period \(t\) .

The lagged variable of growth rate is added to control for any potential endogeneity brought by the omitted variable, in the case when a large influence on current growth by its lagged value is present. Likewise, the lagged value of real GDP per capita is included to test for the convergence effect between high-income and low-income countries. The lagged real GDP per capita refers to the 5-year average of real GDP per capita in the period \(t-1\) .

The variable \({\text{Patents}}_{it}\) stands for the level of AI innovation per country, measured by the total number of AI-related patents per million people in a 5-year average population within period \(t\) . This measure is the same intensity index used by Kim and Lee ( 2015 ) and was also converted into natural logarithms. Footnote 31 AI patents are then replaced by the total number of patents to determine the relationship between total technological innovation and economic growth (see Table 3 ). The expected sign of both patent variables is positive.

\({X}_{it}\) represents a set of control variables that include population growth, real gross capital formation growth rate per capita, real government expenditure growth rate per capita, trade openness, and inflation. Footnote 32 All control variables are 5-year average growth rates except for trade openness, which is the 5-year average ratio of trade volume (exports plus imports) to GDP. Footnote 33 The control variables appear in similar literature, such as in the seminal works of Grier and Tullock ( 1989 ) and Barro ( 1997 ), and in the more recent studies of Bassanini et al. ( 2001 ), Ulku ( 2004 ), Kim et al. ( 2012 ), and Fan and Liu ( 2021 ).

In addition to the control variables, an index using data for years of schooling and returns to education, obtained from the Penn World Table (PWT) by Feenstra et al. ( 2015 ), is taken as a proxy for human capital. Footnote 34 The index makes use of average years of schooling, while also considering decreasing returns to education. Despite this, the index, like other usual human capital measures, ignores cognitive skills, which may be more important in capturing the real effect of human capital (Feenstra et al. 2013 ). This measure also enters the model as a 5-year average growth rate.

Specific time period effects and advanced economic status are indicated using dummy variables. There are ten \(t\) periods in total consisting of five years each, spanning from 1970 to 2019. Advanced economies are countries with more than USD 10,000 of the 5-year average real GDP per capita. Finally, to control for any interaction effect between the level of economic development and patent creation, an interaction term between advanced economic status and patent variables was introduced. The expected sign of the interaction term is negative, implying a lower impact of patent creation on long-run growth among advanced economies.

Statistical treatment was initially done using ordinary least squares (OLS) and fixed effects in panel data. However, because of the inclusion of the lagged growth rate, the model is prone to the Nickell bias, which is unaccounted for in the fixed effects estimation of dynamic panels (Nickell 1981 ; Roodman 2009 ). In addition, bias due to reverse causality between growth rate and patents might be present in the model. Hence, the Anderson-Hsiao (AH) and generalized method of moments (GMM) estimation techniques are employed to minimize endogeneity issues (Arellano and Bond 1991 ; Arellano and Bover 1995 ; Blundell and Bond 1998 ). Footnote 35

The next section presents the results of the panel data regressions.

5 Results and discussion

Table 1 presents the descriptive statistics of the panel data, consisting of ten periods with 5-year intervals between 1970 and 2019. Because of data availability issues, the dataset used is an unbalanced panel data, as indicated by the unequal number of observations ( N ) and number of groups ( n ) across variables.

Five-year growth rates averaged around 1.30%, with a standard deviation of 3.55 across countries in the dataset. Footnote 36 Intuitively, high-income countries will typically have lower growth rates because of the convergence effect. To control for this effect, the estimations presented later include the 5-year average real GDP per capita variable from the previous period. The mean 5-year real GDP per capita is USD 11,990.25.

Table 2 summarizes the economic performance measures such as real GDP per capita and real GDP per capita growth rates, and technological progress in terms of AI and total patents per level of economic development. This follows the classification of Kim and Lee ( 2015 ), where countries with real GDP per capita above USD 10,000 are considered to be in an advanced development stage. Countries with real GDP per capita below the threshold are classified as less developed.

As expected, high-income countries post higher technology output, in terms of patents per million people, between the two income groups. With 165 countries in the dataset, less developed economies are the larger group of the two, and with slightly higher average real GDP per capita growth (1.39%). Illustratively, patent output and income per capita across countries are displayed in Fig.  1 .

AI patents (left), total patents (right), and average real GDP per capita, 1970–2019

Figure  1 depicts the patent publications and level of income per capita. In terms of patents, advanced economies such as Japan, the United States, Germany, South Korea, France, and China have had the highest output between 1970 and 2019. Overall, China has had the highest cumulative AI and total patents within the period, with 849,752 AI and 32,317,932 total patents. This is followed by Japan (365,409 AI and 18,965,778 total patents), and the United States (259,844 AI and 12,883,662 total patents), respectively.

Regardless, all countries started with low levels of AI and total patents in the early 1970s, as illustrated in Fig.  2 . While global AI and total patent counts have steadily increased since the 1970s, China has had a dramatic increase in the number of patents from 2000 onwards. This dwarfs the patent output of other advanced economies (see top panel of Fig.  2 ). The explosion of Chinese patents can be attributed to the growth of R&D expenditure, FDI, and patent subsidies in the country (Chen and Zhang, 2019). Footnote 37

AI patents (left) and total patents (right) of selected countries by period, 1970–2019

Meanwhile, the East Asian economies of Japan and South Korea, have led in terms of patent “intensity,” defined as the number of patents per million people (see bottom panel of Fig.  2 ). Japanese AI and total patents per million people have demonstrated sharp increases since the 1970s but generally declined by the mid-2000s. Footnote 38 On the other hand, South Korea has also witnessed substantial growth in both AI and total patents per million people since the early 1990s. This trend has continued in the subsequent periods, with South Korea eventually overtaking Japan by the early 2010s.

The main estimation results are presented in Table 3 . The estimation techniques used are panel OLS, fixed effects, AH, and GMM. Both models with AI patents and total patents were estimated; columns 1–4 estimate models with the log of AI patents per million people, while columns 5–8 test for the effect of the log of total patents per million people on the dependent variable, denoted by the 5-year average real GDP growth rate. Separate interaction terms between advanced economic status and the patent variables are also included.

As mentioned earlier, the lagged dependent variable in a dynamic panel regression is susceptible to the Nickell bias. Hence, both the AH and GMM estimations are employed to minimize this issue. While the lagged growth rate is positive in all models, it is only statistically significant in the AH estimation among the AI patents models (columns 1–4) and insignificant among the total patents models (columns 5–8). The lack of significance and small magnitude of the coefficients indicate the minimal impact of the first-order lagged growth rate on the contemporaneous growth rate.

Lagged real GDP per capita is statistically significant and negative in all models except in column 8, where it is negative but not significant. The results suggest a strong convergence effect as observed in extant literature. Similarly, population growth has a negative and significant effect on per capita growth in columns 1, 2, 3, and 5, but is insignificant in other estimations. The negative sign of population growth in some estimates is in line with growth theories, but the actual overall effect of population growth across countries is unclear. Kelley and Schmidt ( 1995 ) attribute this mixed result between population growth and economic growth in the long run to the “offsetting” mechanism of “intertemporal demographic effects.” Accordingly, population growth rates are characterized by strong autocorrelation; thus, cross-sectional evidence that uses contemporaneous indicators of population inevitably captures “both the negative impacts of current births and positive impacts of past births.” Footnote 39

Meanwhile, both gross capital formation and government expenditure growth rates manifest significantly positive effects in all models, implying that investments in physical capital and public infrastructure positively contribute to economic growth. Likewise, trade openness is statistically significant and positive in most equations, which is consistent with the existing growth literature. Inflation and human capital, however, are not significant in all models. The lack of significance of human capital can be attributed to (1) the limitations of the measure, and (2) the substitution of labor and/or human capital with AI as an input to production, as raised by Zeira ( 1998 ) and Lu ( 2021 ) among others.

The variables of interest, the extent of AI and total innovation, are taken as the log number of AI patents per million people. As depicted in Table 3 , AI patents have a significant and positive impact on economic growth in all models. This is consistent with the findings of other studies by Lu ( 2021 ) and Yang ( 2022 ). On the other hand, total patents also significantly and positively affect economic growth; however, the magnitude of the effect is lower than AI patents. This is somewhat consistent with the findings of Nguyen and Doytch ( 2022 ), wherein total patents do not display a significant impact on the long-run growth rate. In addition, Nguyen and Doytch ( 2022 ) conclude that ICT patents have a more significant impact on economic growth than other kinds of patents. Footnote 40

To address potential endogeneity either by omitted variables or reverse causality, the patent variables are instrumented in GMM estimations (columns 4 and 8). Footnote 41 Estimated coefficients of AI innovation seem to be consistent and significant at least at the 10% level. On the other hand, total innovation is significant at least at the 10% level in OLS, fixed effects, and AH, but insignificant in GMM.

Moreover, the level of economic development (advanced economy) variable is insignificant in all estimated models except in column 6. Meanwhile, the interaction term between the level of economic status and patent creation is negative and significant in columns 4 and 6. The negative sign implies that AI and overall innovation exhibit less impact on economic growth among advanced economies, which is similar to the convergence effect stated previously.

Furthermore, the Sargan-Hansen test provides the test for overidentifying restrictions for the GMM model. The p-values of the Sargan-Hansen statistic of the GMM models for AI and total patents are 0.426 and 0.581, respectively. Thus, the null hypothesis that the instruments are valid is not rejected. Footnote 42 Also, the Arellano-Bond AR(1) and AR(2) tests for GMM are presented for reference. The p-values of the AR tests indicate the presence of serial correlation only at the first differences. Footnote 43

Results indicate that AI-related innovation drives long-run economic growth. The wide applicability of AI across industries can be one reason for its positive contribution. AI systems can be implemented in manufacturing, ICT, transportation, finance, and medical services among other industries (Mou 2019 ). Self-learning and monitoring benefit the manufacturing sector by increasing precision and efficient utilization of physical capital, reducing defects and delays (Rao and Verweij 2017 ). More recent and practical forms of AI such as voice-to-text applications and speech recognition allow businesses to reach and respond to customers in real time (Mou 2019 ), inducing an increase in the volume of consumer transactions. AI technologies can be implemented in financial systems to detect fraudulent activities, preventing theft and loss (Bose 2006 ; Akhilomen 2013 ). Furthermore, predictive modeling with AI can analyze and manage traffic flow (Mou 2019 ; Yigitcanlar et al. 2020 ), which is notoriously known to cause negative externalities, more effectively.

5.1 Robustness checks

For additional robustness checks, periodic estimations in the dataset are also performed. The dataset is divided into two periods, 1970–1994 and 1995–2019, consisting of 25 years each. Due to limited AI and non-AI patent data, the 1970–1994 period has fewer observations. Additionally, most countries that published and applied for AI-related patents within this period are advanced economies, as shown in the number of groups ( n ) and the lack of an estimated coefficient for the “advanced” dummy in columns 2 and 3 in Table 4 . Footnote 44 Therefore, a comparison of impacts on long-run growth brought by technological innovation, specifically those on AI, between advanced and less advanced economies might not be intuitively useful for observations within this time frame.

Table 4 displays the estimation results of the models for both AI and total patents for the 1970–1994 period, whereas Table 5 provides the results of estimations for the period 1995–2019. As indicated in Table 4 , the effect of AI on growth is not statistically significant for the period 1970–1994, which can be due to (1) the limited number of observations, (2) the lesser number of AI patents, and (3) the relatively less technically advanced nature of AI innovation during this period. Interestingly, the effect of total patents is significant and positive during the same period, as shown in columns 5–8. Hence, the results suggest that other types of patents compared to early AI technologies might have had a more substantial effect on growth rates before 1995.

On the other hand, the estimated fixed effects, AH, and GMM coefficients are significant for AI patents in the 1995–2019 period in columns 2–4 in Table 5 . The significance of the estimates provides evidence of AI being a driver of long-run economic growth for the latter half of the time frame in the dataset. More surprisingly, the value of the coefficient of AI patents in the GMM model is relatively large compared to estimated parameters in other models. Meanwhile, the total patents variable is insignificant for long-run growth rate in all models, except in OLS in column 5. In addition, the interaction terms between patent creation and economic status are mostly insignificant in both periods. Hence, there is no clear distinction on the effect of patent creation between developed and less developed economies on long-run growth in both periods. Footnote 45

An obvious implication of the above results is that the effect of AI has become increasingly evident toward the latter years of the dataset, while innovations from other disciplines have extended relatively less impact on growth. Footnote 46 Notably, AI patent registration had started picking up by the mid-1990s, especially among advanced economies (see Fig.  2 ), which naturally contributed to a heightened impact of AI in the second half of the entire period. Arguably, the quality of AI technologies within this period has also improved compared to earlier forms of AI prior to the last two to three decades.

Furthermore, separate estimations between advanced and less advanced economies were also done, both for AI and total patents. As defined in the previous section, advanced countries are those with more than USD 10,000 of the 5-year average real GDP per capita. The results can be found in both Tables 6 and 7 .

The effect of AI is strongly and positively significant among advanced economies in columns 1–4 in Table 6 , but does not entail any implication on long-run growth among less advanced or emerging economies in Table 7 . Footnote 47 This suggests that viable infrastructures and institutions, which may only be available in developed countries, are necessary to leverage AI in the economy. This, in turn, translates into positive contributions to economic growth. More importantly, this finding resembles the theory proposed by Zeira ( 1998 ), which explains the differences in the type and level of technologies available across countries.

Meanwhile, total patents engender a quite similar effect on growth between advanced and less advanced economies. Coefficients of the patent variable are positive and significant in OLS, but negligible in fixed effects, AH, and GMM, which is true for both groups of countries. This indicates that total patents do not contribute to long-run economic growth regardless of a country's level of development. Thus, more specific, technical, and practical innovations, such as those of AI or ICT in nature, are more important than other types of innovations in terms of their effect on economic growth.

Finally, the possibility of an external instrument is not precluded and has been explored to further address endogeneity. As mentioned earlier, the estimated model is susceptible to bias, either due to omitted variables or bi-directional causality between patent creation and economic growth. Hence, aside from fixed effects and GMM, fixed effects estimation with instrumental variable (FE-IV) is also considered as a means of obtaining unbiased estimates.

The number of non-patent literature (NPL) cited by the patents is used as an instrument for both AI and total patents. NPL refers to the cited articles of a patent document that are not patents themselves (e.g., scientific publications, books, online sources, conference proceedings, etc.) to “justify” an invention’s “novelty.” Furthermore, NPL citations help gauge “the impact of scientific production cited in patents,” or conversely, “the technological impact of scientific publications” (Velayos-Ortega and Lopez-Carreño 2021 ).

Non-patent references contribute to patent creation by providing justification and a scientific foundation for the technology being patented. Hence, a rich amount of non-patent knowledge should positively contribute to patent creation. Scientific knowledge itself is vast and varied; however, only the cited literature in the patents themselves is specific and relevant to the inventions.

As expected, the direct effect of scientific publications, as a measure of scientific knowledge, on economic growth has been well-studied in the literature (e.g., Kim and Lee 2015 ; Solarin and Yen 2016 ; Maradana et al. 2017 ; Pinto and Teixeira 2020 ). While the non-significance of academic research on economic growth has been found in some studies, general sentiment and results still regard academic publications as a direct and positive contributor to growth. This notion casts some doubt on whether scientific literature can serve as a valid instrument for patent creation.

This study, however, suggests that for research output to be translated into an object of economic value, it has to be transformed first into an input (or intermediate good), to be used later in the production of other goods. Footnote 48 The knowledge contained in relevant and cited NPL, for example, is used by patent creators, or inventors, to create new products, services, modes of production, processes, frameworks, and/or other kinds of inventions used for enterprise building. Thus, the transformation of scientific knowledge into intermediate, technology-based capital goods is embodied in the patents themselves. Finally, the high patent output indicates the availability of technology that helps in the production of final goods, which then ultimately leads to economic growth. Footnote 49

The results of the FE-IV regression are shown in Table 8 , alongside the OLS, fixed effects, and GMM estimations. Due to the limited data on the instruments, the number of observations and groups in Table 8 is lower compared to the number of observations and groups in the main results (see Table 3 ). Footnote 50 For convenience, the same panel groups used in the FE-IV regression are used in the OLS, fixed effects, and GMM estimations as well to allow comparison of the estimates.

The log number of AI patents per million people is significantly positive in all models (at the 10% level in GMM and FE-IV), and the magnitudes of the AI coefficients are relatively consistent among the fixed effects, GMM, and FE-IV estimations in columns 2, 3, and 4 in Table 8 , respectively. Notably, the magnitudes of the coefficients are larger than the estimates in the main results in Table 3 . On the other hand, total patents are only significant and positive in OLS and fixed effects in columns 5 and 6. In addition, the magnitude of the coefficient of total patents in the FE-IV regression (column 8) is inconsistent with the other estimations.

Several tests were performed to check for the validity of the instruments in both the GMM and FE-IV models. The null hypothesis is not rejected for the Sargan-Hansen test for overidentifying restrictions in both the GMM and FE-IV estimations, suggesting no overidentification in the first-stage regressions. This is true for both the AI and total patents models (columns 3, 4, 7, and 8 in Table 8 ). Meanwhile, the null hypothesis of the Kleibergen-Paap test for weak instruments is rejected for the FE-IV estimates of both the AI and total patents models, implying the first stage FE-IV regressions are not underidentified. Hence, both tests seem to confirm the validity of the instruments used, especially for the FE-IV estimations.

The p-values of the endogeneity test, however, differ between the AI patents and total patents models in FE-IV (columns 4 and 8). Under the null hypothesis, the regressors, or the instruments, can be treated as exogenous. Rejection of the null hypothesis means the regressors are not exogenous and thus may not be considered acceptable instruments. According to the test, the null hypothesis is not rejected for AI patents but is rejected for the total patents model at standard significance levels. The result indicates that the validity of instruments and parameter estimates is only applicable to the former, whereas estimates for the latter model are likely inconsistent and biased.

Overall, the results of the main estimations and robustness checks reveal a strong positive relationship between AI innovation and long-run economic growth. This is consistent with the endogenous growth theories and with the findings of existing literature such as Kim and Lee ( 2015 ), He ( 2019 ), and Nguyen and Doytch ( 2022 ). On the other hand, total patents still contribute to long-run economic growth, albeit to a lesser extent compared to more technical innovations such as those developed in ICT. This is particularly true to more recent observations in the dataset. Moreover, AI has had a more robust and significant effect on the long-run growth among advanced economies, while total innovation exhibits almost no impact on growth for both advanced and emerging countries.

In addition, an IV estimation with fixed effects using cited NPL has been employed to further minimize the endogeneity issue. The FE-IV estimates are valid for the AI patents model, but not for the total patents model. The FE-IV estimates are also comparable with the results of other estimation techniques such as fixed effects and GMM, suggesting that cited non-patent references may serve as an instrument for specific types of patents such as those related to AI.

6 Conclusion

Innovations in AI have been around as early as the 1970s, but their application and impact have only been more apparent and pervasive in the last ten to twenty years. The huge surge in AI and total patent registrations by the turn of the century, alongside the obvious physical and non-physical embodiments of innovative technologies consumed daily, is evidence of how AI and related technologies have changed the economic landscape. Several companies, especially those in e-commerce, have been leveraging natural language processing to predict customer behavior to increase sales. Meanwhile, multinational companies rely on AI and machine learning to optimize their supply chains through predictive scheduling, demand forecasting, inventory and risk management, and predictive maintenance among many other purposes (Rao and Verweij 2017 ; Ashcroft 2023 ). In general, advancements in AI and related ICT technologies have ultimately helped in modernizing production processes, minimizing manual inefficiency, and enhancing overall customer experience across firms and industries.

This study sets out to determine the relationship between the level of AI innovation and long-run economic growth, using a panel dataset across countries between 1970 and 2019. The main finding demonstrates that there exists a positive and significant impact of AI patenting on average long-run economic growth. Additionally, the effect of AI is more apparent in the latter period, because of the increasing quantity and quality of AI innovation generated over time. Overall, the positive impact of AI found in this study is consistent with the results of other studies focusing on AI and growth such as those by He ( 2019 ), Fan and Liu ( 2021 ), and Yang ( 2022 ).

Meanwhile, there is also some evidence of the positive contribution of total patent creation on economic growth. This positive effect of patenting is consistent with the findings of Wong et al. ( 2005 ), Kim and Lee ( 2015 ), and Nguyen and Doytch ( 2022 ). The effect, however, is notably smaller and weaker compared to the effect of AI patents on growth. Total patents, however, have exhibited significantly positive effects in the earlier periods of the dataset. The muted effect of patent publication on long-run economic growth is similar to the results found by Chu et al. ( 2016 ), Blind et al. ( 2022 ), and Nguyen and Doytch ( 2022 ) in their studies.

Furthermore, the effect of AI on growth is more robust among advanced economies, which is in line with the theory of machine automation proposed by Zeira ( 1998 ). Because of differences in capital endowments, not all countries can keep up with the pace of a constantly shifting technological frontier. As AI requires physical, and oftentimes ICT-related capital and technical know–how, not all countries can implement and use AI technologies effectively. In the meantime, more developed economies can leverage AI in production and business operations because of the availability of knowledge and infrastructure that complement AI, which engenders a strong positive contribution of AI to economic growth.

Finally, cited non-patent references in AI patents may serve as a valid instrument for AI patent creation. The estimates obtained from the FE-IV regression are consistent with the fixed effects estimation and GMM, and are also supported by various tests on instrument validity. Further work on this topic is recommended to future researchers, either by discovering other possible instruments or expanding the use of the instrument to other types of patents and/or measures of innovation.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. However, the datasets supporting the conclusions of this article are also available in the following repositories: Google Patents Data: https://console.cloud.google.com/bigquery?ws=!1m4!1m3!3m2!1spatents-public-data!2sgoogle_patents_research . Penn World Table: https://www.rug.nl/ggdc/productivity/pwt/?lang=en . United Nations Statistics: https://unstats.un.org/unsd/snaama/Downloads . World Bank Development Indicators: https://databank.worldbank.org/source/world-development-indicators .

Yang et al. ( 2021 ), Mahalakshmi et al. ( 2022 ), Sood et al. ( 2022 ), etc.

Zeira ( 1998 ) notes that standard economic growth models that involve technology adoption encourage the accumulation of capital. However, this may not be necessary as “new technology increases output for any combination of inputs”.

Arrow ( 1962 ) assumes a competitive equilibrium. However, Dasgupta and Stiglitz ( 1988 ) reason that learning possibilities can only translate into growth if “learning spillovers are complete.” In an oligopolistic market structure, for example, firms may not be able to learn “costlessly, completely, and instantaneously from the experience of others.”.

Arrow ( 1962 ) follows this idea from the standard neoclassical growth model. On the contrary, Bahk and Gort ( 1993 ) revealed that the effect of learning can also be disembodied from both capital and labor.

Zeira ( 1998 ) ascribes this to the wage differential across countries. Countries with lower wages tend to have lower productivity; hence, they are unable to afford the high requirements of capital to adopt the latest technology.

Low-indexed tasks refer to tasks that require minimal skill. In general, low-indexed tasks are assigned to low-skilled labor.

Acemoglu and Restrepo ( 2018 ) attribute the stability of the growth path to “self-correcting forces” of the factor prices. Furthermore, as both types of innovation advance at the same rate, the long-run growth rate path is characterized by a constant labor share.

This study also suffers from this problem. For example, the chosen measure for AI may still not fully and accurately capture its effect on growth.

See Aghion and Howitt ( 1990 ).

Chu et al. ( 2016 ), however, claim that “patent breadth” only promotes growth in the short run by raising the “profit margin of monopolistic firms” and providing “more incentives for R&D.” Accordingly, patents reduce growth in the long run but expand the total number of firms. Further, they state that an R&D subsidy is a more appropriate policy for “stimulating long-run economic growth.”.

Kim and Lee ( 2015 ), however, recognize that this might only be plausible among advanced countries, where viable “national innovation systems” and infrastructures enable scientific research to be put into effective commercial use.

This is similar to the finding of Kim et al. ( 2012 ), where they found no significant effect of patent intensity on growth among developing economies.

In relation to this, Lee et al. ( 2011 ) highlight a “mutual causation” between research publications and GDP among Asian economies. This causation, however, is less clear in Western countries.

Mueller ( 2006 ) distinguishes the types of R&D (private and university) from each other and estimated the impact of each separately. Also, the university-industry relation was measured by industry grants per researcher.

In the sample, Nakatani ( 2021 ) reveals an insignificant impact of intangible assets on TFP among South Korean firms, whereas a significant and positive contribution to TFP with the largest magnitude is found for ICT firms in the United Kingdom.

Analytical AI refers to technologies that help in the identification of “new insights, patterns, and relationships or dependencies” in data for decision-making. Functional AI executes or implements actions, instead of generating recommendations. Interactive AI enables “interactive communication” between the user and a smart system to provide user assistance (e.g., chatbots and smart personal assistants). Textual AI typically covers textual analytics and natural language processing. Finally, visual AI can be considered a “branch of computer science” that “trains” machines to learn images and visual data (Sarker 2022 ).

The period considered in the analysis may not cover recent generative AI such as ChatGPT, unless these inventions have been patented years before their release.

Aside from these challenges, training AI models is typically associated with enormous costs, both in time and resources.

Challenges in measurement gauge how well an AI performs its assigned tasks. On the other hand, generalization and calibration refer to the performance and replicability of an AI system to different populations or datasets. Local context encompasses the “act of fitting” new technology and its “goodness of fit” into a pre-existing organizational network (Coiera 2019 ).

This is similar to the Bahk and Gort ( 1993 ) model. Lu ( 2021 ) further adds that AI may replace human labor in the future, which subsequently has welfare implications.

The balanced growth path by Lu ( 2021 ) shows output, human capital, physical capital, AI, and consumption grow at the same rate.

Specifically, He ( 2019 ) measures AI as “the ratio of fixed assets investment in information transmission computer services and software industry to GDP.”.

Fan and Liu ( 2021 ) have developed an index to measure AI level based on three aspects, namely “infrastructure development, technology application, and market benefits.”.

On the other hand, Oxford Insights ( 2022 ) has developed an AI readiness index per country, available in the annual reports published since 2017. However, the current dataset lacks enough observation in terms of the time dimension. Thus, using the index was ruled out in favor of long-run analysis. Nonetheless, this study recommends using the index and/or other related AI measures for future research once more data are available.

Jones and Williams ( 1998 ) explored the idea of a non-constant return to \(R\) and \(A\) . They introduce additional parameters that represent “congestion externality,” “knowledge spillovers,” and “fishing out effects” in research, allowing the parameter \(\theta\) to fall between 0 and 1 and assume non-linearity in \(A\) .

Equation (6) is the (short-run) growth equation when \(dt=1\) . For long-run growth rates, the change in \(t\) is greater than 1 \((dt>1)\) to indicate longer periods.

Griliches ( 1998 ), however, acknowledges the difficulty of measuring these relationships, as knowledge stock is unobservable.

Lu and Zhou ( 2021 ) note that the definition of AI in theoretical models can be “very broad,” whereas empirical data tend to have “narrow” definitions, resulting in a gap between the two. Theoretical models typically depict AI as a type of automation, but continuous AI development may be capable of replacing even high-skilled labor. In addition, AI raises the question of what a “human being” is in economics, where the human being is often “narrowed down” to “labor” and an “optimization agent.” Aside from the current lack of clarity of whether AI is a “new production technology” or simply a new input of production, the question of which input (e.g., labor, human capital, or an “independent decision-making agent”) is AI used as a substitute for also persists.

In addition, Brynjolfsson et al. ( 2018 ) highlighted the “modern productivity paradox” in the age of AI. AI is indeed capable of many promising feats; however, productivity growth remained stagnant over the past decade. They attributed this inconsistency to several reasons, such as the difficulty of measuring AI capital because of its mostly intangible outputs, and the amount of time and resources required for the impact of technology to be fully reflected in productivity.

Because of the mode of data extraction, the AI patent variable may be prone to accuracy and measurement error. As much as possible, the list of common AI terms used in the text search has been exhaustive. Furthermore, some technical jargon may be shared among multiple branches of knowledge that include AI. Hence, the word list has been limited to the most common and specific AI terms. An exact search of the identified terms and/or phrases was then performed.

Several studies make use of R&D “intensity” as a measure of innovation (e.g. Jones and Williams 1998 ; Blind et al. 2006 ; Yanhui et al. 2015 ). Other examples of R&D intensity measures include patent applications per R&D expenditure, R&D over sales for firm-level data, number of researchers per million people, etc.

Because of data availability, this study makes use of the implicit price deflator (rather than the consumer price index) to calculate inflation.

Except for trade, all variables in \({X}_{it}\) are also expressed as instantaneous growth rates.

This comes from Wößmann ( 2003 ), who argues that common proxies for human capital such as school enrollment rates and average years of schooling either insufficiently or incorrectly model the “development effect” of human capital. Specifically, Wößmann ( 2003 ) explains that enrollment ratios are flow variables, and enrolled students are not yet part of the labor force, and thus are excluded from economic production. On the other hand, average years of schooling “misspecifies” human capital by placing the “same weight on any year of schooling” of a person, and does not input the “quality of education system.”.

The full list of variables is available in Table 12 in the appendix. Additional variables are considered (e.g., Internet users, non-patent literature) as part of the robustness checks. See Sect.  5.1 under Sect. 5 .

The presence of positive and negative outliers contributed to a relatively high standard deviation. Calculated five-year real GDP per capita growth rates range between − 24.52% and 23.59% in the dataset, across countries and periods.

The driving forces, however, have had specific and varying effects per type of patent filing. Chen and Zhang ( 2019 ) note that R&D spending generally boosts Chinese patent creation, while FDI is only robust for utility and design patents. Patent subsidies, on the other hand, have a positive effect on design patents.

The gradual decrease in domestic patenting was due to Japanese firms being selective in their patent registrations, focusing more on the quality than the number of filings (Japan Patent Office 2015 ).

Kelley and Schmidt ( 1995 ), however, note that this does not imply that demographic effects on per capita growth are unimportant. Empirical results only highlight the need to study the long-run dynamics between population growth and output growth more carefully.

Aside from the technology “intensity index” given by the log number of patents per million people, the log number of patents was also used directly in the estimations. Results of these estimations reveal similar results (positive and significant coefficient for AI patents, and weak significance for total patents).

The period (1970–2019) considered for estimation covers several socio-economic, political, and technological events (e.g., military conflicts, oil shocks, financial crises, Internet diffusion, etc.) that may have affected inter-country growth rates. All estimated models include time dummies; however, they may not fully capture the influence of external events on long-run growth rates. As part of the robustness checks, the average five-year growth rate of Internet users per country, for example, was included as a control variable. Results are available in Table 10 in the appendix.

The reliability of the Sargan-Hansen statistic, however, weakens as the number of instruments increases. Thus, the number of instruments was reduced to avoid this issue as much as possible. Roodman ( 2009 ) recommends that the total number of instruments should be less than the total number of individual units in a panel dataset. To reduce the number of instruments, the optimal number of lags is chosen per GMM estimation. A maximum of five lags is used, but the model should simultaneously satisfy the Sargan-Hansen and AR(2) tests, while also considering the explanatory power of the variable(s) of interest and control variables. Following these specifications, the AI patents model passes the Sargan-Hansen and AR(2) tests until the fifth-order lagged instruments, while the total patents model only passes both tests at the second lag; hence, the difference in the number of instruments (column 4 with 65 and column 8 with 32).

Unit root tests (Fisher-type based on augmented Dickey-Fuller) for unbalanced panel data were also performed to check for random walk. The tests revealed that the variables contain at least one stationary panel (the null hypothesis that all n panels contain unit roots is rejected). However, the test could not be performed for the log of AI patents per million people due to missing observations.

The “advanced” dummy variable was dropped due to collinearity.

A short-run analysis was also conducted to check for the short-run impacts of technical innovation on economic growth. Instead of a five-year average growth rate, the yearly real GDP growth rate per capita growth rate was used as the dependent variable. Likewise, annual levels and growth rates of the patent and control variables were used in the regressions. The results are available in Table 9 in the appendix.

Distinguishing the effect of AI from non-AI patents on growth might be a challenging task when using patents as an indicator of technological innovation. As AI becomes increasingly and deeply embedded in production tools and processes, any new invention might have some AI component in it. Hence, disembodying AI from the “non-AI” component of an invention, for example, to estimate AI’s true effect on growth might present a challenge for future research.

Because of the limited number of countries, the number of instruments (42) used in the GMM estimation among advanced economies for AI patents is relatively close to the number of individual panels n (53). While the number of instruments is still lower than the number of individual groups, it is more desirable to have the number of instruments as few as possible.

Although Pinto and Teixeira ( 2020 ) use research output instead of patents as a measure of knowledge as a good, the authors illustrate how research ultimately contributes to economic growth (see Fig.  1 of Pinto and Teixeira 2020 ).

Moreover, to strengthen the exogeneity assumption, only the NPL cited in the patents themselves is used as an instrument, rather than the entire population of scientific and academic publications. Hence, the instrument used has a direct causal link to patent creation and is more likely to manifest an effect on growth only through the patents.

Both the current and lagged values of the (natural log) number of non-patent literature are used as instruments to ensure the validity and precision of the FE-IV estimates. Also, the interaction term between advanced economic status and patent variable is instrumented. The first stage results are available in Table 11 in the appendix.

Acemoglu D, Restrepo P (2018) The race between man and machine: implications of technology for growth, factor shares, and employment. Am Econ Rev 108(6):1488–1542. https://doi.org/10.1257/aer.20160696

Article   Google Scholar  

Aghion P, Jones BF, Jones CI (2018) Artificial intelligence and economic growth. In: The economics of artificial intelligence: an agenda, pp. 237–282. University of Chicago Press. https://doi.org/10.3386/w23928

Aghion P, Howitt P (1990) A model of growth through creative destruction. Econometrica 60(2):323–351. https://doi.org/10.2307/2951599

Akhilomen J (2013) Data mining application for cyber credit-card fraud detection system. In: Industrial Conference on Data Mining, pp. 218–228. https://doi.org/10.1007/978-3-642-39736-3_17

Arellano M, Bond S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev Econ Stud 58(2):277–297. https://doi.org/10.2307/2297968

Arellano M, Bover O (1995) Another look at the instrumental variable estimation of error-components models. J Econometr 68(1):29–51. https://doi.org/10.1016/0304-4076(94)01642-D

Arrow KJ (1962) The economic implications of learning by doing. Rev Econ Stud 29(3):155–173. https://doi.org/10.2307/2295952

Ashcroft S. Top 10 AI and ML supply chain firms & solutions.” Supply Chain Digital , May 24. 2023; Available at https://supplychaindigital.com/digital-supply-chain/top-10-ai-and-ml-supply-chain-solutions

Bahk B-H, Gort M (1993) Decomposing learning by doing in new plants. J Polit Econ 101(4):561–583. https://doi.org/10.1086/261888

Barro RJ (1997) Determinants of economic growth: a cross-country empirical study. MIT Press, Cambridge. https://doi.org/10.3386/w5698

Book   Google Scholar  

Bassanini A, Stefano S, Philip H (2001) Economic growth: the role of policies and institutions. Panel data evidence from OECD countries. In: OECD Economics Department Working Papers, No. 283, OECD Publishing, Paris. https://doi.org/10.1787/722675213381 .

Blind K, Edler J, Frietsch R, Schmoch U (2006) Motives to patent: empirical evidence from Germany. Res Policy 35(5):655–672. https://doi.org/10.1016/j.respol.2006.03.002

Blind K, Ramel F, Rochell C (2022) The influence of standards and patents on long-term economic growth. J Technol Transf 47(4):979–999. https://doi.org/10.1007/s10961-021-09864-3

Blundell R, Bond S (1998) Initial conditions and moment restrictions in dynamic panel data models. J Econometr 87(1):115–143. https://doi.org/10.1016/S0304-4076(98)00009-8

Bose R (2006) Intelligent technologies for managing fraud and identity theft. In: Third International Conference on Information Technology: New Generations (ITNG'06), pp. 446–451. https://doi.org/10.1109/ITNG.2006.78 .

Brynjolfsson E, Daniel R, Chad S (2018) Artificial intelligence and the modern productivity paradox: a clash of expectations and statistics. The economics of artificial intelligence: An agenda, pp. 23–57. University of Chicago Press. https://doi.org/10.3386/w24001 .

Chen Z, Zhang J (2019) Types of patents and driving forces behind the patent growth in China. Econ Model 80:294–302. https://doi.org/10.1016/j.econmod.2018.11.015

Chu AC, Furukawa Y, Ji L (2016) Patents, R&D subsidies, and endogenous market structure in a Schumpeterian economy. South Econ J 82(3):809–825. https://doi.org/10.1002/soej.12122

Coiera E (2019) The last mile: where artificial intelligence meets reality. J Med Internet Res 21(11):e16323. https://doi.org/10.2196/16323

Dasgupta P, Stiglitz J (1988) Learning-by-doing, market structure and industrial and trade policies. Oxf Econ Pap 40(2):246–268. https://doi.org/10.1093/oxfordjournals.oep.a041850

Fan D, Liu K (2021) The relationship between artificial intelligence and China’s sustainable economic growth: focused on the mediating effects of industrial structural change. Sustainability 13(20):11542. https://doi.org/10.3390/su132011542

Feenstra RC, Robert I, Marcel T (2015) The next generation of the Penn world table. Am Econ Rev 105(10):3150–3182. https://doi.org/10.1257/aer.20130954

Feenstra RC, Robert I, Marcel T (2013) PWT 8.0—a user guide. Available at: www.ggdc.net/pwt .

Google (2022) “Google Patents Public Data.” Available at: https://console.cloud.google.com/bigquery?ws=!1m4!1m3!3m2!1spatents-public-data!2sgoogle_patents_research

Grier KB, Tullock G (1989) An empirical analysis of cross-national economic growth, 1951–1980. J Monet Econ 24(2):259–276. https://doi.org/10.1016/0304-3932(89)90006-8

Griliches Z (1988) Patent statistics as economic indicators: a survey. R&D Productiv Econ Evid pp. 287–343. University of Chicago Press. https://doi.org/10.3386/w3301 .

Hatemi-J A, Ajmi AN, El Montasser G, Inglesi-Lotz R, Gupta R (2016) Research output and economic growth in G7 countries: new evidence from asymmetric panel causality testing. Appl Econ 48(24):2301–2308. https://doi.org/10.1080/00036846.2015.1117052

He Y (2019) The effect of artificial intelligence on economic growth: evidence from cross-province panel data. Korea J Artif Intell 7(2):9–12. https://doi.org/10.24225/kjai.2019.7.2.9

Inglesi-Lotz R, Balcilar M, Gupta R (2014) Time-varying causality between research output and economic growth in US. Scientometrics 100(1):203–216. https://doi.org/10.1007/s11192-014-1257-z

Institut national de la statistique et des études économiques (INSEE). 2019. “Technologies de l'information et de la communication et commerce électronique (TIC) 2019.” Available at: https://www.insee.fr/fr/metadonnees/source/fichier/TIC2019_questionnaire.pdf

International Telecommunication Union (2023) Artificial intelligence for good. Available at: https://www.itu.int/en/mediacentre/backgrounders/Pages/artificial-intelligence-for-good.aspx .

Japan Patent Office (2015) JPO Status Report 2015. Available at: https://www.jpo.go.jp/e/resources/report/statusreport/2015/ .

Jones CI, Williams JC (1998) Measuring the social return to R&D. Q J Econ 113(4):1119–1135. https://doi.org/10.2139/ssrn.2155

Kelley AC, Schmidt RM (1995) Aggregate population and economic growth correlations: the role of the components of demographic change. Demography 32(4):543–555. https://doi.org/10.2307/2061674

Kim YK, Lee K (2015) Different impacts of scientific and technological knowledge on economic growth: contrasting science and technology policy in East Asia and Latin America. Asian Econ Policy Rev 10(1):43–66. https://doi.org/10.1111/aepr.12081

Kim YK, Lee K, Park WG, Choo K (2012) Appropriate intellectual property protection and economic growth in countries at different levels of development. Res Policy 41(2):358–375. https://doi.org/10.1016/J.RESPOL.2011.09.003

Lach S (1995) Patents and productivity growth at the industry level: a first look. Econ Lett 49(1):101–108. https://doi.org/10.1016/0165-1765(94)00618-C

Lee L-C, Lin P-H, Chuang Y-W, Lee Y-Y (2011) Research output and economic productivity: a Granger causality test. Scientometrics 89(2):465–478. https://doi.org/10.1007/s11192-011-0476-9

Lu C-H (2021) The impact of artificial intelligence on economic growth and welfare. J Macroecon 69:103342. https://doi.org/10.1016/j.jmacro.2021.103342

Lu Y, Zhou Y (2021) A review on the economics of artificial intelligence. J Econ Surv 35(4):1045–1072. https://doi.org/10.1111/joes.12422

Mahalakshmi V, Nirzar Kulkarni KV, Pradeep Kumar K, Suresh Kumar D, Sree N, Durga S (2022) The role of implementing artificial intelligence and machine learning technologies in the financial services industry for creating competitive intelligence. Mater Today Proc 56:2252–2255. https://doi.org/10.1016/j.matpr.2021.11.577

Maradana RP, Pradhan RP, Dash S, Gaurav K, Jayakumar M, Chatterjee D (2017) Does innovation promote economic growth? Evidence from European countries. J Innov Entrep 6(1):1–23. https://doi.org/10.1186/s13731-016-0061-9

Marr B (2023) A short history of chatGPT: how we got to where we are today. Forbes, May 19. Available at: https://www.forbes.com/sites/bernardmarr/2023/05/19/a-short-history-of-chatgpt-how-we-got-to-where-we-are-today/?sh=1b6986bf674f

Montagnier P, Irene EK (2021) AI measurement in ICT usage surveys: a review. OECD Digital Economy Papers, No. 308, OECD Publishing, Paris. https://doi.org/10.1787/20716826 .

Mou X (2019) Artificial intelligence: investment trends and selected industry uses. International Finance Corporation 8. Available at: https://openknowledge.worldbank.org/handle/10986/32652

Mueller P (2006) Exploring the knowledge filter: How entrepreneurship and university–industry relationships drive economic growth. Res Policy 35(10):1499–1508. https://doi.org/10.1016/J.RESPOL.2006.09.023

Nakatani R (2021) Total factor productivity enablers in the ICT industry: a cross-country firm-level analysis. Telecommun Policy 45(9):102188. https://doi.org/10.1016/j.telpol.2021.102188

Nchake MA, Shuaibu M (2022) Investment in ICT infrastructure and inclusive Growth in Africa. Sci Afr 17:1–16. https://doi.org/10.1016/j.sciaf.2022.e01293

Nguyen CP, Nadia D (2022) The impact of ICT patents on economic growth: an international evidence. Telecommun Policy 46(5):102291. https://doi.org/10.1016/j.telpol.2021.102291

Nickell S (1981) Biases in dynamic models with fixed effects. Econometrica 49(6):1417–1426. https://doi.org/10.2307/1911408

Nightingale P (2000) Economies of scale in experimentation: knowledge and technology in pharmaceutical R&D. Ind Corp Chang 9(2):315–359. https://doi.org/10.1093/ICC/9.2.315

Ntuli H, Inglesi-Lotz R, Chang T, Pouris A (2015) Does research output cause economic growth or vice versa? Evidence from 34 OECD countries. J Am Soc Inf Sci 66(8):1709–1716. https://doi.org/10.1002/asi.23285

Oxford Insights (2022) Government AI Readiness Index 2022. Available at: https://www.oxfordinsights.com/government-ai-readiness-index-2022 .

Pinto T, Teixeira AAC (2020) The impact of research output on economic growth by fields of science: a dynamic panel data analysis, 1980–2016. Scientometrics 123(2):945–978. https://doi.org/10.1007/s11192-020-03419-3

Rao AS, Gerard V (2017) Sizing the prize: What’s the real value of AI for your business and how can you capitalise. PwC Publication, PwC: 1–30. Available at: https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf

Romer PM (1990) Endogenous technological change. J Polit Econ 98(5):S71–S102. https://doi.org/10.1086/261725

Roodman D (2009) How to do xtabond2: an introduction to difference and system GMM in Stata. Stand Genomic Sci 9(1):86–136. https://doi.org/10.1177/1536867X0900900106

Sarker IH (2022) Ai-based modeling: techniques, applications and research issues towards automation, intelligent and smart systems. SN Comput Sci 3(2):158. https://doi.org/10.1007/s42979-022-01043-x

Sinha D (2008) Patents, innovations and economic growth in Japan and South Korea: evidence from individual country and panel data. Appl Econ Int Dev 8(1). Available at: https://ssrn.com/abstract=1308261

Solarin SA, Yen YY (2016) A global analysis of the impact of research output on economic growth. Scientometrics 108(2):855–874. https://doi.org/10.1007/s11192-016-2002-6

Sood SK, Keshav SR, Dheeraj K (2022) A visual review of artificial intelligence and Industry 4.0 in healthcare. Comput Electr Eng 101:107948. https://doi.org/10.1016/j.compeleceng.2022.107948

Statistics Sweden (2020) Artificial intelligence (AI) in Sweden 2019. Available at: https://www.scb.se/en/finding-statistics/statistics-by-subject-area/education-and-research/research/research-and-development-in-sweden/pong/statistical-news/artificial-intelligence-ai-in-sweden-2019/

Sweet C, Eterovic D (2019) Do patent rights matter? 40 years of innovation, complexity and productivity. World Dev 115:78–93. https://doi.org/10.1016/J.WORLDDEV.2018.10.009

Szczepański M (2019) Economic impacts of artificial intelligence (AI). European Parliamentary Research Service. Available at: https://www.europarl.europa.eu/RegData/etudes/BRIE/2019/637967/EPRS_BRI(2019)637967_EN.pdf

The European Commission (2021) Community Survey on ICT Usage and E-Commerce in Enterprises. Available at: https://circabc.europa.eu/sd/a/f9dc8b66-a429-49e2-ae01-f7424ec389f0/MQ_2021_ICT_ENT.pdf

The United Nations (2022) National Accounts—analysis of Main Aggregates (AMA). https://unstats.un.org/unsd/snaama/Downloads

The World Bank (2022) World Development Indicators. The World Bank Group. https://databank.worldbank.org/source/world-development-indicators

Trajtenberg M (2018) AI as the next GPT: a Political-Economy Perspective. No. w24245. National Bureau of Economic Research. https://doi.org/10.3386/w24245 .

Ulku H (2004) R&D, innovation, and economic growth: an empirical analysis. Int Monet Fund Working Papers 185:2–35. https://doi.org/10.5089/9781451859447.001

Velayos-Ortega G, López-Carreño R (2021) Non-patent literature. Encyclopedia 1(1):198–205. https://doi.org/10.3390/encyclopedia1010019

Wang L, Zhan J, Shi W, Liang Yi (2011) In cloud, can scientific communities benefit from the economies of scale? IEEE Trans Parallel Distrib Syst 23(2):296–303. https://doi.org/10.1109/TPDS.2011.144

Wong PK, Ho YP, Autio E (2005) Entrepreneurship, innovation and economic growth: evidence from GEM data. Small Bus Econ 24(3):335–350. https://doi.org/10.1007/S11187-005-2000-1

Wößmann L (2003) Specifying human capital. J Econ Surv 17(3):239–270. https://doi.org/10.1111/1467-6419.00195

Yang C-H (2022) How artificial intelligence technology affects productivity and employment: firm-level evidence from Taiwan. Res Policy 51(6):104536. https://doi.org/10.1016/j.respol.2022.104536

Yang T, Yi X, Shaowen Lu, Johansson KH, Chai T (2021) Intelligent manufacturing for the process industry driven by industrial artificial intelligence. Engineering 7(9):1224–1230. https://doi.org/10.1016/j.eng.2021.04.023

Yanhui W, Huiying Z, Jing W (2015) Patent elasticity, R&D intensity and regional innovation capacity in China. World Patent Inf 43:50–59. https://doi.org/10.1016/J.WPI.2015.10.003

Yeung K (2020) Recommendation of the council on artificial intelligence (OECD). Int Leg Mater 59(1):27–34. https://doi.org/10.1017/ilm.2020.5

Yigitcanlar T, Desouza KC, Butler L, Roozkhosh F (2020) Contributions and risks of artificial intelligence (AI) in building smarter cities: insights from a systematic review of the literature. Energies 13(6):1473. https://doi.org/10.3390/en13061473

Zeira J (1998) Workers, machines, and economic growth. Q J Econ 113(4):1091–1117. https://doi.org/10.1162/003355398555847

Download references

Acknowledgements

The Article Processing Charge was covered by the funds of PAPAIOS and JSPS (KAKENHI Grant Number JP 21HP2002). The author wishes to acknowledge Craig Parsons, who provided valuable assistance and knowledge in the authorship of this manuscript.

Declarations

The author confirms to have read, understood, and agreed to the submission guidelines, policies, and submission declaration of the journal. The author confirms that the manuscript is the author's original work and that the manuscript has not received prior publication and is not under consideration for publication elsewhere. The author confirms that the paper now submitted is not copied or plagiarized version of some other published work.

The author received no financial support for the research, authorship, and/or publication of this article.

Author information

Authors and affiliations.

Yokohama National University, 79-1 Tokiwadai, Hodogaya Ward, Yokohama, Kanagawa, 240-0067, Japan

Julius Tan Gonzales

You can also search for this author in PubMed   Google Scholar

Contributions

Not applicable.

Corresponding author

Correspondence to Julius Tan Gonzales .

Ethics declarations

Competing interests.

The author confirms that the manuscript has no competing interest to declare. The author declares that there are no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

See Tables 9 , 10 , 11 , 12 .

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Gonzales, J.T. Implications of AI innovation on economic growth: a panel data study. Economic Structures 12 , 13 (2023). https://doi.org/10.1186/s40008-023-00307-w

Download citation

Received : 02 June 2023

Revised : 27 August 2023

Accepted : 28 August 2023

Published : 09 September 2023

DOI : https://doi.org/10.1186/s40008-023-00307-w

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Artificial intelligence
• Economic growth

			, and No 23928, from This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but A.I. now seems poised to automate many tasks once thought to be out of reach, from driving cars to making medical recommendations and beyond. How will this affect economic growth and the division of income between labor and capital? What about the potential emergence of “singularities” and “superintelligence,” concepts that animate many discussions in the machine intelligence community? How will the linkages between A.I. and growth be mediated by firm-level considerations, including organization and market structure? The goal throughout is to refine a set of critical questions about A.I. and economic growth and to contribute to shaping an agenda for the field. One theme that emerges is based on Baumol’s “cost disease” insight: growth may be constrained not by what we are good at but rather by what is essential and yet hard to improve. (search for similar items in EconPapers) 2017-10 this item is included in EFG PR (100) as Artificial Intelligence and Economic Growth , Philippe Aghion, Benjamin F. Jones, Charles I. Jones. in The Economics of Artificial Intelligence: An Agenda , Agrawal, Gans, and Goldfarb. 2019 (external link) (application/pdf) Chapter: (2018) for items with the same title. BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text This working paper can be ordered from for this paper in NBER Working Papers from National Bureau of Economic Research, 1050 Massachusetts Avenue Cambridge, MA 02138, U.S.A.. Contact information at . Bibliographic data for series maintained by ( ). Academia.edu no longer supports Internet Explorer. To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser . Enter the email address you signed up with and we'll email you a reset link. We're Hiring! Help Center Artificial Intelligence and Economic Growth: A Theoretical Framework Scientific Annals of Economics and Business The growing adoption of Artificial Intelligence (AI) has sparked ubiquitous concerns worldwide. Artificial intelligence can affect economic growth and employment. The influence is assumed to be substantial because the adoption of AI technology may lead to increased productivity, lower wages, prices, and labor substitution. Artificial intelligence can affect global economic growth with its widespread adoption and diffusion. We mathematically examined the effects of AI on economic growth, reiterating how AI is unique as a production factor. The models show that AI capital lowers capital prices, increases wages, and augments productivity. Besides, AI capital positively affects the labor share and vice versa, provided that AI and labor are complementary. We improved a task-based model to show AI raises both labor share and wages by generating new tasks. We also present the potential policy implications of AI adoption. We conclude AI can contribute to economic growth. Labor-abundant coun... Related Papers Economie et Statistique / Economics and Statistics Céline Antonin Julius Tan Gonzales The application of artificial intelligence (AI) across firms and industries warrants a line of research focused on determining its overall effect on economic variables. As a general-purpose technology, for example, AI helps in production, marketing, and customer acquisition of firms, increasing their productivity and consumer reach. Aside from these, other effects of AI include enhanced quality of services, improved work accuracy and efficiency, and increased customer satisfaction. Hence, this study aims to gauge the impact of AI on the economy, specifically on long-run economic growth. This study conjectures a significantly positive relationship between AI and economic growth. To test this hypothesis, this study makes use of a panel dataset of countries from 1970 to 2019, and the number of AI patents as measure of AI. A text search query is performed to distinguish AI patents from other types of innovations in a public database. Employing fixed effects and GMM estimation, this pape... SSRN Electronic Journal Yingying Lu International Journal of Sustainable Economies Management ASHFAQ AHAMED A The general purpose of this scientific paper (Paper) is to analyze the impact of artificial intelligence (AI) and digitalization on the macroeconomics of the EU member states (including Romania). To reach the general objective, 6 specific objectives were selected: the analysis of the birth and evolution of the concept of AI; identification and analysis of various forms of artificial intelligence, of its possible impact on the aggregate economy, based on the findings of the scientific writings; identification of the relevant sources of statistic data and information enabling us to undertake an inter-state comparative study regarding AI and digitalization; analysis of the intensity and sense of the correlation between various indicators expressing the use of AI systems and digitalization, in the EU; analysis of the interdependence between the average GDP per capita and labor productivity on the one hand, and the AI and digitalization levels, on the other; analysis of the interdependen... Syed Adnan Jawaid Background: The swift assimilation of Artificial Intelligence (AI) across multiple industries is altering the worldwide economic terrain. Unprecedented opportunities and difficulties are brought about by this revolutionary technology, which has an impact on innovation, trade, labor markets, and security. Objective: The goal of this study is to thoroughly investigate the economic effects of AI by exploring how it may affect labor markets, innovation, trade internationally, and security. The goal is to offer insights that support strategic decision-making and policy formation by examining current trends, opportunities, and difficulties. Results: The research highlights the dual character of artificial intelligence's influence, highlighting both its potential to spur economic expansion and its drawbacks, including the loss of jobs, a lack of standards, and security issues. The study highlights how crucial cooperation is to overcoming these obstacles and realizing AI's full potential. Conclusion: The balancing opportunities and risks becomes critical as AI continues to change the economic landscape. The report promotes proactive steps including ethical AI design guidelines, ongoing worker retraining, and flexible regulatory strategies. As we shape an AI future that puts inclusion, creativity, and responsible governance first, collaboration emerges as a key theme. Ajay Agrawal Lianzi Zhou IZA Journal of Labor Policy Ekkehard Ernst The current wave of technological change based on advancements in artificial intelligence (AI) has created widespread fear of job loss and further rises in inequality. This paper discusses the rationale for these fears, highlighting the specific nature of AI and comparing previous waves of automation and robotization with the current advancements made possible by a widespread adoption of AI. It argues that large opportunities in terms of increases in productivity can ensue, including for developing countries, given the vastly reduced costs of capital that some applications have demonstrated and the potential for productivity increases, especially among the low skilled. At the same time, risks in the form of further increases in inequality need to be addressed if the benefits from AI-based technological progress are to be broadly shared. For this, skills policies are necessary but not sufficient. In addition, new forms of regulating the digital economy are called for that prevent fur... Eurasian Business Review Giacomo Damioli Recent evidence indicates an upsurge in artificial intelligence and robotics (AI) patenting activities in the latest years, suggesting that solutions based on AI technologies might have started to exert an effect on the economy. We test this hypothesis using a worldwide sample of 5257 companies having filed at least a patent related to the field of AI between 2000 and 2016. Our analysis shows that, once controlling for other patenting activities, AI patent applications generate an extra-positive effect on companies’ labor productivity. The effect concentrates on SMEs and services industries, suggesting that the ability to quickly readjust and introduce AI-based applications in the production process is an important determinant of the impact of AI observed to date. Loading Preview Sorry, preview is currently unavailable. You can download the paper by clicking the button above. RELATED PAPERS Science, Technology and Development Dr. G. YOGANANDHAM Journal of Economics, Finance and Accounting Studies JAMIELYN BONSAY Sema Yılmaz Genç giuseppe ruggeri Computer Science & Information Technology (CS & IT) Austine O . Unuriode , Alexander T . Lopez , Mayowa J Akinwande , Stanley Okoro International Journal of Computational Intelligence Systems Dr.Leonardus W W Mihardjo Human Systems Management MOK AL AWADI IEEE Xplore Dr. Gajendra Kumar Gupta Jo Ann Oravec The 6th International Conference on Economics and Social Sciences Gabriela Călinescu Anayochukwu K I N G S L E Y Ugwu ICPA Journal of Business Administration and Economics Prof. Onah Vitalis Chukwuma ResearchGate Abu Rayhan , Swajan Rayhan Importance of Artificial Intelligence (AI) in the Economy (Atena Editora) Atena Editora Ayisha Tabbassum Proceedings of the National Academy of Sciences Esteban Moro Seth Benzell Ed Fernandez Strategic Imperatives and Core Competencies in the Era of Robotics and Artificial Intelligence Robert Niewiadomski Journal of Liberty and International Affairs, Institute for Research and European Studies - Bitola Kwadwo Bonsu Jonathan Kolber   We're Hiring!   Help Center Find new research papers in: Health Sciences Earth Sciences Cognitive Science Mathematics Computer Science Academia ©2024 Artificial Intelligence and Economic Growth NBER Working Paper No. w23928 57 Pages Posted: 16 Oct 2017 Last revised: 20 May 2023 Philippe Aghion College de France and London School of Economics and Political Science, Fellow; Centre for Economic Policy Research (CEPR); National Bureau of Economic Research (NBER) Benjamin F. Jones Northwestern University ; National Bureau of Economic Research (NBER) Charles I. Jones Stanford Graduate School of Business; National Bureau of Economic Research (NBER) Date Written: October 2017 This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but A.I. now seems poised to automate many tasks once thought to be out of reach, from driving cars to making medical recommendations and beyond. How will this affect economic growth and the division of income between labor and capital? What about the potential emergence of “singularities” and “superintelligence,” concepts that animate many discussions in the machine intelligence community? How will the linkages between A.I. and growth be mediated by firm-level considerations, including organization and market structure? The goal throughout is to refine a set of critical questions about A.I. and economic growth and to contribute to shaping an agenda for the field. One theme that emerges is based on Baumol’s “cost disease” insight: growth may be constrained not by what we are good at but rather by what is essential and yet hard to improve. Suggested Citation: Suggested Citation Philippe Aghion (Contact Author) College de france and london school of economics and political science, fellow ( email ). London United Kingdom Centre for Economic Policy Research (CEPR) National bureau of economic research (nber). 1050 Massachusetts Avenue Cambridge, MA 02138 United States Northwestern University ( email ) 2001 Sheridan Road Evanston, IL 60208 United States 847-491-3177 (Phone) 847-467-1777 (Fax) Stanford Graduate School of Business ( email ) Stanford GSB 655 Knight Way Stanford, CA 94305-4800 United States 650-725-9265 (Phone) HOME PAGE: http://www.stanford.edu/~chadj Do you have a job opening that you would like to promote on SSRN? Paper statistics, related ejournals, entrepreneurship, innovation, & growth ejournal. Subscribe to this fee journal for more curated articles on this topic Political Economy - Development: Domestic Development Strategies eJournal National bureau of economic research (nber) series. Subscribe to this free journal for more curated articles on this topic My Favorites You have successfully logged in but... ... your login credentials do not authorize you to access this content in the selected format. Access to this content in this format requires a current subscription or a prior purchase. Please select the WEB or READ option instead (if available). Or consider purchasing the publication. OECD Artificial Intelligence Papers The impact of artificial intelligence on productivity, distribution and growth. The OECD engages in comprehensive research and analysis to deepen insights into the transformative power of artificial intelligence (AI) and its implications for economies and societies. The OECD Artificial Intelligence Papers series draws on comprehensive evidence-base to identify trends and developments and delve into an extensive array of AI-related subjects. This includes opportunities and challenges in AI governance, the impact of AI across policy domains including labour markets, education and healthcare, ethical considerations in AI design and deployment, and new frontiers such as generative AI. This collection encompasses OECD policy papers as well as working papers. Science and Technology Social Issues/Migration/Health ISSN: 3005348X (online) https://doi.org/10.1787/dee339a8-en Subscribe to the RSS feed Subscribe to the RSS feed Key mechanisms, initial evidence and policy challenges This paper explores the economics of Artificial Intelligence (AI), focusing on its potential as a new General-Purpose Technology that can significantly influence economic productivity and societal wellbeing. It examines AI's unique capacity for autonomy and self-improvement, which could accelerate innovation and potentially revive sluggish productivity growth across various industries, while also acknowledging the uncertainties surrounding AI's long-term productivity impacts. The paper discusses the concentration of AI development in big tech firms, uneven adoption rates, and broader societal challenges such as inequality, discrimination, and security risks. It calls for a comprehensive policy approach to ensure AI's beneficial development and diffusion, including measures to promote competition, enhance accessibility, and address job displacement and inequality. Click to access: Click to download PDF - 1.93MB PDF Cite this content as: Author(s) Francesco Filippucci, Peter Gal, Cecilia Jona-Lasinio, Alvaro Leandro and Giuseppe Nicoletti 16 Apr 2024 DOI: 10.3386/W23928 Corpus ID: 26299957 Artificial Intelligence and Economic Growth Philippe Aghion , Benjamin F. Jones , Charles I. Jones Published in The Economics of Artificial… 1 October 2017 Economics, Computer Science Kauffman: Conferences & Seminars (Topic) 357 Citations The importance of artificial intelligence to economic growth, will this time be different a review of the literature on the impact of artificial intelligence on employment, incomes and growth. Highly Influenced Artificial Intelligence, Jobs, Inequality and Productivity: Does Aggregate Demand Matter? The impact of artificial intelligence on economic development, explosive growth from ai automation: a review of the arguments, artificial intelligence as a service, economic growth, and well-being, the complex economics of artificial intelligence, artificial intelligence and worker power, 1 artificial intelligence and its implications for income distribution and unemployment 1, artificial intelligence and entrepreneurship: opportunities and challenges, 74 references, intelligence explosion microeconomics, a time to sow and a time to reap: growth based on general purpose technologies, the time scale of artificial intelligence: reflections on social effects, the race between man and machine: implications of technology for growth, factor shares, and employment. Highly Influential Capital Accumulation and Economic Growth Are we approaching an economic singularity information technology and the future of economic growth, american economic association macroeconomics of unbalanced growth : the anatomy of urban crisis, factor-eliminating technical change, economics for the common good, research and development in the growth process, related papers. Showing 1 through 3 of 0 Related Papers Search Menu Sign in through your institution Browse content in Arts and Humanities Browse content in Archaeology Biblical Archaeology Historical Archaeology Prehistoric Archaeology Browse content in Architecture History of Architecture Browse content in Art History of Art Theory of Art Browse content in Classical Studies Classical History Classical Philosophy Classical Reception Classical Literature Classical Art and Architecture Greek and Roman Archaeology Religion in the Ancient World Browse content in History Colonialism and Imperialism Environmental History Genocide and Ethnic Cleansing Historical Geography History by Period History of Education History of Gender and Sexuality Industrial History Intellectual History International History Labour History Legal and Constitutional History Maritime History Military History National Liberation and Post-Colonialism Oral History Political History Public History Regional and National History Revolutions and Rebellions Slavery and Abolition of Slavery Social and Cultural History Theory, Methods, and Historiography Urban History World History Browse content in Linguistics Applied Linguistics Historical and Diachronic Linguistics Phonetics and Phonology Sociolinguistics Browse content in Literature Literary Studies (European) Literary Studies (Eco-criticism) Literary Studies (Romanticism) Literary Studies (American) Literary Studies - World Literary Studies (1500 to 1800) Literary Studies (19th Century) Literary Studies (20th Century onwards) Literary Studies (African American Literature) Literary Studies (British and Irish) Literary Studies (Early and Medieval) Literary Studies (Fiction, Novelists, and Prose Writers) Literary Studies (Gender Studies) Literary Studies (History of the Book) Literary Studies (Plays and Playwrights) Literary Studies (Poetry and Poets) Literary Studies (Postcolonial Literature) Literary Studies (Queer Studies) Literary Studies (Science Fiction) Literary Studies (War Literature) Literary Studies (Women's Writing) Literary Theory and Cultural Studies Shakespeare Studies and Criticism Browse content in Media Studies Browse content in Music Applied Music Dance and Music Ethnomusicology Gender and Sexuality in Music Music Cultures Music and Media Music and Culture Music Theory and Analysis Musical Structures, Styles, and Techniques Musicology and Music History Race and Ethnicity in Music Browse content in Philosophy Aesthetics and Philosophy of Art Epistemology Feminist Philosophy History of Western Philosophy Metaphysics Moral Philosophy Philosophy of Science Philosophy of Language Philosophy of Mind Philosophy of Religion Philosophy of Mathematics and Logic Practical Ethics Social and Political Philosophy Browse content in Religion Biblical Studies Christianity History of Religion Judaism and Jewish Studies Religion and Politics Religion and Art, Literature, and Music Religious Studies Browse content in Society and Culture Cultural Studies Ethical Issues and Debates Technology and Society Visual Culture Browse content in Law Browse content in Company and Commercial Law Commercial Law Company Law Browse content in Comparative Law Systems of Law Competition Law Browse content in Constitutional and Administrative Law Government Powers Local Government Law Military and Defence Law Parliamentary and Legislative Practice Criminal Law Employment and Labour Law Environment and Energy Law Browse content in Financial Law History of Law Human Rights and Immigration Intellectual Property Law Browse content in International Law Public International Law IT and Communications Law Jurisprudence and Philosophy of Law Law and Politics Law and Society Browse content in Legal System and Practice Courts and Procedure Medical and Healthcare Law Browse content in Policing Criminal Investigation and Detection Terrorism and National Security Law Browse content in Medicine and Health History of Medicine Browse content in Public Health and Epidemiology Public Health Browse content in Science and Mathematics Browse content in Biological Sciences Aquatic Biology Biochemistry Bioinformatics and Computational Biology Ecology and Conservation Evolutionary Biology Genetics and Genomics Microbiology Molecular and Cell Biology Natural History Plant Sciences and Forestry Research Methods in Life Sciences Zoology and Animal Sciences Browse content in Earth Sciences and Geography Maps and Map-making Palaeontology Urban Geography Browse content in Engineering and Technology Biological Engineering Environmental Science History of Science and Technology Browse content in Mathematics Pure Mathematics Browse content in Physics Mathematical and Statistical Physics Browse content in Psychology Social Psychology Browse content in Social Sciences Browse content in Anthropology Anthropology of Religion Medical Anthropology Physical Anthropology Regional Anthropology Social and Cultural Anthropology Theory and Practice of Anthropology Browse content in Business and Management Business History Corporate Governance Industry Studies Organizational Theory and Behaviour Criminology and Criminal Justice Browse content in Economics Agricultural, Environmental, and Natural Resource Economics Asian Economics Behavioural Economics and Neuroeconomics Econometrics and Mathematical Economics Economic Systems Economic History Economic Methodology Economic Development and Growth Financial Markets Financial Institutions and Services Health, Education, and Welfare History of Economic Thought International Economics Labour and Demographic Economics Macroeconomics and Monetary Economics Microeconomics Public Economics Urban, Rural, and Regional Economics Welfare Economics Browse content in Education Early Childhood and Elementary Education Educational Strategies and Policy Higher and Further Education Philosophy and Theory of Education Secondary Education Browse content in Human Geography Political Geography Interdisciplinary Studies Browse content in Politics Comparative Politics Environmental Politics European Union Gender and Politics International Relations Political Methodology Political Sociology Political Behaviour Political Economy Political Institutions Political Theory Politics and Law Public Policy US Politics Browse content in Regional and Area Studies African Studies Asian Studies Browse content in Sociology Comparative and Historical Sociology Economic Sociology Gender and Sexuality Gerontology and Ageing Health, Illness, and Medicine Marriage and the Family Migration Studies Occupations, Professions, and Work Organizations Population and Demography Race and Ethnicity Social Theory Social Movements and Social Change Social Research and Statistics Social Stratification, Inequality, and Mobility Sociology of Religion Sociology of Education Sport and Leisure Urban and Rural Studies Journals on Oxford Academic Books on Oxford Academic < Previous chapter Next chapter > 9 Artificial Intelligence and Economic Growth Published: May 2019 Cite Icon Cite Permissions Icon Permissions This chapter examines the potential impact of artificial intelligence (AI) on economic growth. We model AI as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but AI now seems poised to automate many tasks once thought to be out of reach, from driving cars to making medical recommendations and beyond. How will this affect economic growth and the division of income between labor and capital? What about the potential emergence of "singularities" and "superintelligence," concepts that animate many discussions in the machine-intelligence community? How will the linkages between AI and growth be mediated by firm-level considerations, including organization and market structure? The goal throughout is to refine a set of critical questions about AI and economic growth, and to contribute to shaping an agenda for the field. One theme that emerges is based on Baumol's "cost disease" insight: growth may be constrained not by what we are good at but rather by what is essential and yet hard to improve. Personal account Sign in with email/username & password Get email alerts Save searches Purchase content Activate your purchase/trial code Add your ORCID iD Institutional access Sign in with a library card. Sign in with username/password Recommend to your librarian Institutional account management Get help with access Access to content on Oxford Academic is often provided through institutional subscriptions and purchases. If you are a member of an institution with an active account, you may be able to access content in one of the following ways: IP based access Typically, access is provided across an institutional network to a range of IP addresses. This authentication occurs automatically, and it is not possible to sign out of an IP authenticated account. Choose this option to get remote access when outside your institution. Shibboleth/Open Athens technology is used to provide single sign-on between your institution’s website and Oxford Academic. Click Sign in through your institution. Select your institution from the list provided, which will take you to your institution's website to sign in. When on the institution site, please use the credentials provided by your institution. Do not use an Oxford Academic personal account. Following successful sign in, you will be returned to Oxford Academic. If your institution is not listed or you cannot sign in to your institution’s website, please contact your librarian or administrator. Enter your library card number to sign in. If you cannot sign in, please contact your librarian. Society Members Society member access to a journal is achieved in one of the following ways: Sign in through society site Many societies offer single sign-on between the society website and Oxford Academic. If you see ‘Sign in through society site’ in the sign in pane within a journal: Click Sign in through society site. When on the society site, please use the credentials provided by that society. Do not use an Oxford Academic personal account. If you do not have a society account or have forgotten your username or password, please contact your society. Sign in using a personal account Some societies use Oxford Academic personal accounts to provide access to their members. See below. A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions. Some societies use Oxford Academic personal accounts to provide access to their members. Viewing your signed in accounts Click the account icon in the top right to: View your signed in personal account and access account management features. View the institutional accounts that are providing access. Signed in but can't access content Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access. If you believe you should have access to that content, please contact your librarian. For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more. Our books are available by subscription or purchase to libraries and institutions. Month: Total Views: October 2022 1 November 2022 7 December 2022 5 January 2023 4 February 2023 4 March 2023 3 April 2023 5 June 2023 5 July 2023 4 October 2023 5 November 2023 9 December 2023 2 January 2024 6 February 2024 14 March 2024 9 April 2024 7 May 2024 7 June 2024 2 July 2024 2 August 2024 5 About Oxford Academic Publish journals with us University press partners What we publish New features   Open access Rights and permissions Accessibility Advertising Media enquiries Oxford University Press Oxford Languages University of Oxford Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide Copyright © 2024 Oxford University Press Cookie settings Cookie policy Privacy policy Legal notice This Feature Is Available To Subscribers Only Sign In or Create an Account This PDF is available to Subscribers Only For full access to this pdf, sign in to an existing account, or purchase an annual subscription. Artificial Intelligence and Economic Growth This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but A.I. now seems poised to automate many tasks once thought to be out of reach, from driving cars to making medical recommendations and beyond. How will this affect economic growth and the division of income between labor and capital? What about the potential emergence of "singularities" and "superintelligence," concepts that animate many discussions in the machine intelligence community? How will the linkages between A.I. and growth be mediated by firm-level considerations, including organization and market structure? The goal throughout is to refine a set of critical questions about A.I. and economic growth and to contribute to shaping an agenda for the field. One theme that emerges is based on Baumol's "cost disease" insight: growth may be constrained not by what we are good at but rather by what is essential and yet hard to improve. Related Topics Working Paper More Publications The productivity of professions: evidence from the emergency department, granular search, market structure, and wages, the minimum wage and the market for low-skilled labor: why a decade can make a difference. Architecture and Design Asian and Pacific Studies Business and Economics Classical and Ancient Near Eastern Studies Computer Sciences Cultural Studies Engineering General Interest Geosciences Industrial Chemistry Islamic and Middle Eastern Studies Jewish Studies Library and Information Science, Book Studies Life Sciences Linguistics and Semiotics Literary Studies Materials Sciences Mathematics Social Sciences Sports and Recreation Theology and Religion Publish your article The role of authors Promoting your article Abstracting & indexing Publishing Ethics Why publish with De Gruyter How to publish with De Gruyter Our book series Our subject areas Your digital product at De Gruyter Contribute to our reference works Product information Tools & resources Product Information Promotional Materials Orders and Inquiries FAQ for Library Suppliers and Book Sellers Repository Policy Free access policy Open Access agreements Database portals For Authors Customer service People + Culture Journal Management How to join us Working at De Gruyter Mission & Vision De Gruyter Foundation De Gruyter Ebound Our Responsibility Partner publishers Your purchase has been completed. Your documents are now available to view. 9. Artificial Intelligence and Economic Growth From the book the economics of artificial intelligence. Philippe Aghion , Benjamin F. Jones and Charles I. Jones X / Twitter Supplementary Materials Please login or register with De Gruyter to order this product. Chapters in this book (31) nb Norwegian Will the AI Revolution Lead to Greater Prosperity? Artificial intelligence has the potential to reshape our economies, labor markets, societies, and politics. But despite the rosy forecasts of an AI-driven boom, history shows that technological advances rarely lead to immediate improvements in living standards and often lead to profound disruption. CAMBRIDGE – As global economic growth slows, many hope technological innovation is a potential solution. The International Monetary Fund’s latest World Economic Outlook , for example, highlighted the potential of artificial intelligence to boost productivity and GDP. But the report also warns that given the uncertainties surrounding the extent of AI’s impact, such forecasts should be approached with a dose of caution. While AI could usher in an era of prosperity, this outcome depends on how these emerging technologies evolve. To continue reading, register now. It’s free! Register Now Already have an account? Log in ✕ Log in/Register Please log in or register to continue. Registration is free and requires only your email address. Email required Password required Remember me? Please enter your email address and click on the reset-password button. If your email exists in our system, we'll send you an email with a link to reset your password. Please note that the link will expire twenty-four hours after the email is sent. If you can't find this email, please check your spam folder. Reset Password Cancel PS Economics Newsletter PS Politics Newsletter Marketing Communications & Updates By proceeding, you are agreeing to our Terms and Conditions . Sign in with Your Institution Edit Newsletter Preferences Set up notification. To receive email updates regarding this {entity_type}, please enter your email below. If you are not already registered, this will create a PS account for you. You should receive an activation email shortly. Latest News Cryptocurrencies White Papers Maximizing Economic Impact: How to Fine-Tune AI Artificial intelligence has grown to become a force in the global economy; its wide possibility is for unprecedented growth, efficiency, and innovation. The optimization of the described AI systems for maximum economic impact must be about adjusting the AI technologies to ensure effective productivity and growth and address some of humanity's pressing problems. Here is a detailed explanation of how to tune AI for maximum economic impact: Realizing the Economic Potential of AI AI can contribute a lot to increasing the global economic output. According to McKinsey , AI may add up to $13 trillion to the world economy by 2030, hence representing a potential augmentation of global GDP by approximately 1.2% annually. These potentials result from automation of tasks, better decision-making, and the forming of new products and services brought about by AI. This encompasses the task of developing and tuning true strategic AI technologies. Key Areas of AI Optimization 1. Improvement of Productivity: Artificial Intelligence can improve productivity through the automation of more mundane tasks, hence putting workers to do what they were meant for and detailing their focus toward doing complicated and creative activities. This would include AI-powered tools that handle activities such as data entry, customer service inquiries, and even areas of decision-making. Fine-tuning the algorithms of AI, in trying to understand and predict human behavior, would obviate many operations, hence cutting down on many costs for businesses. 2. Accelerating Innovation: AI will act as a catalyst for innovation in terms of new products and new services. For example, AI can look over large data sets—not so much with human eye recognition, but, rather, it could discover trends and insights. Then, many innovations in the health and financial sectors may well be associated with it. Proper tuning of the analytical capabilities of AI will make a business stronger than others and therefore a driver of the economy. 3. Improved Decision Making: AI enables decision-making powered by accurate and on-time insights. Machine learning algorithms learn from past data and predict future trends, making businesses undertake suitable decisions. Allowing its interpolation to make it more accurate and reliable may result in improvement, such as in the fields of supply chain management, marketing, and financial planning. 4. Tackling Real-World Societal Wicked Problems —AI has the solutions for some of the most pressing problems humanity faces today, such as ensuring good health, affordable and quality healthcare, providing excellent education, and surviving in the age of climate change. For example, AI can be used in patient-centered treatment plans, the personalization of education for learners, and monitoring changes in the natural environment. Tuning an AI for these challenges can gain enormous benefits. Fine-Tuning AI Strategies 1. Quality and Quantity of Data: However, AI systems are only as good as the quality and quantity of the data they have been subjected to provide them with training. In line with this, AI algorithms fundamentally need data of very high quality for accurate learning and reliable predictions. In this regard, businesses need to ensure that their AI systems get the best data by investing in input collection and management processes. This can be realized by increasing the quantity of data, which boosts performance in artificial intelligence models. 2. Algorithm Optimization: Automated hyperparameter optimization may be employed in the optimization of the whole fine-tuning algorithms to enhance performance. This translates virtually into tuning an algorithm for the outcome. Considerations for businesses, therefore, may need to take a notch up with more sophisticated techniques of machine learning, like in cases of deep-learning techniques or reinforcement learning to optimize or fine-tune the potential or the capacity of their AI systems. 3. Continuous Learning and Adaptation: The AI system should be designed to learn from experience and improve itself over time, specifically with mechanisms for continued training and updating of AI models. Such continuous re-adjustment will assist a business in keeping AI systems in tune and effective in a changing environment. 4. Ethical Considerations: AI systems should be tuned subject to ethical considerations. Companies must verify that their AI technologies work in a transparent, fair, and accountable manner. This care will involve questions about bias, privacy, and security. Early built-in ethical considerations in the design and development of AI systems can boost trust between the relevant stakeholders and the maximum benefit of AI possible. 5. Collaboration and Knowledge Sharing- Collaboration and knowledge sharing are two elements, crucial to ensure that the possible economic benefits of AI are realized. He stated, that the sharing of best practices and the development of new AI technologies with academic institutions, government agencies, and other organizations through detailed collaboration will result in a robust AI ecosystem that transits into innovation and hence results in economic growth. Case Studies 1. Health Sector: AI technology has found its application in the health sector by developing predictive models for disease diagnosis and treatment. For instance, AI algorithms can make analyses of the medical imagery to reveal the first signs of cancer. The algorithms can be tuned to be sharper and more accurate, hence allowing health providers to extend better care with lower spending. 2. Finance Area: Applying to the finance area, AI is used for fraud detection, risk management, and personalized financial services. For example, AI-based chatbots can advise the customer about financial matters by their spending patterns. If such AI-enabled systems come to understand the customer behavioral pattern much better, there are chances of even increasing revenue with dramatically better service. 3. Retail: AI is used in retail for inventory management, personalizing marketing campaigns, and therefore enhancing the customer shopping experience. With the help of AI algorithms dedicated to the analysis of customer data, they could forecast the demand for a product while, at the same time, having the ability to adjust their levels of stock. Retailers can fine-tune these to reduce wastage and improve efficiency while maximizing sales. 4. Manufacturing - It finds an application in manufacturing for the optimization of processes, controlling quality, and reducing the downtime of the manufacturing process. For example, AI-based predictive maintenance systems track how efficient the machines are and deduce when maintenance might be required next. The implications are cost savings and productivity enhancement, achieved by fine-tuning such a system for higher accuracy. Problems and Solutions 1. Data Privacy and Security: One of the major challenges in tuning AI is privacy and security in data. Enterprises should develop strong measures for protecting the data and making sure that no data that should be leaked to unwanted stakeholders is taken care of. Proper encryption techniques, access control mechanisms, and carrying out regular security audits are important. 2. Bias and Fairness - Some AI systems are designed inherently such that the result from them is biased at times and hence provide unfair results. In this regard, businesses should include bias detection and techniques to alleviate bias in AI algorithms. These techniques may range from the use of several training datasets to frequent audits and integrating fairness metrics into the assessment of the AI model. 3. Skill Gaps - Demand for professionals who are competent in developing and fine-tuning AI systems is growing and will necessarily mean investment in employee development programs by the business for the creation of a relevantly skilled workforce through AI and machine learning courses, workshops, and certifications. 4. Regulatory Compliance - Proper AI functioning requires compliance; therefore businesses have to keep up to date with changes in laws and regulations pertinent to the operation of the business. Such worry entails data protection, ethical guidelines, and industry standards The Need to Make AI More Accurate. In this reverence only, businesses can leverage the maximum potential of AI by optimizing their AI systems toward maximum productivity, innovation, decisions, and resolutions for societal needs. This will call for such things as being oriented around data quality, optimized algorithms, continuous learning, ethical considerations, and collaboration, among others, as per the strategies. Consequently, AI will be a great tool in the churning of economic growth and, in essence, making the future perfect if the right strategies are used. Related Stories MIT Technology Review Newsletters What will AI mean for economic inequality? If we’re not careful, we could see widening gaps within countries and between them. Katya Klinova archive page Prominent AI researchers expect the arrival of artificial general intelligence anywhere between “the next couple of years” and “possibly never.” At the same time, leading economists disagree about the potential impact of AI: Some anticipate a future of perpetually accelerating productivity, while others project more modest gains. But most experts agree that technological advancement, however buoyant, is no guarantee that everyone benefits.  And unfortunately, even though some of the most notable AI R&D efforts declare that making sure everyone benefits is a key goal or guiding principle, ensuring that AI helps create a more inclusive future remains one of the least invested-in areas of AI governance. This might seem natural given the state of the field: The impact AI will have on labor and inequality is still highly uncertain, making it difficult to design interventions. But we know at least some of the factors that will influence the interplay between AI and inequality over the next few decades. Paying attention to those can help us make the idea that AI will benefit everyone into more than just a pipe dream. Because they’re largely driven by the private sector, AI development and use are heavily influenced by the incentive structures of the world’s economies. And if there is something important that can be predicted with reasonable certainty about those economies, it is their future demographic composition. There is a stark divide between higher-income countries, whose populations are aging rapidly and will shrink without migration, and low- and lower-middle income countries, which will continue to grow for the rest of the century thanks to the excess of births over deaths. What does this have to do with AI? AI development is concentrated in the aging countries, and thus it will follow the path set by the realities, needs, and incentives in those places. Aging countries are seeing the ratio of working-age people to retirees collapse, making it more difficult to sustain pension schemes and contain health-care costs. Countries looking to maintain their retirees’ living standards and their overall economic dynamism will seek ways to expand their effective labor force, be that with humans or with artificial agents. Limited (and likely highly unpopular) gains could come from increasing the retirement age. More sizable gains could come from immigration. But keeping the ratio of the working-age to retiree populations constant would require a significant increase in immigration to the higher-income countries. Widespread anti-immigration sentiment makes that seem unlikely, though opinions could change relatively quickly when people are faced with the prospect of diminishing pensions and rising health-care costs.   If overly restrictive immigration policies do not relax in rich countries, we will likely see the economic incentives to fill labor gaps with AI go into overdrive over the next few decades. It might seem on the surface that this won’t exacerbate inequality if there are fewer people than available jobs. But if the trend is associated with an uneven distribution of gains and losses, increasingly precarious employment, excessive surveillance of workers, and digitization of their know-how without adequate compensation, we should expect a spike in inequality.  And even if the efforts to replace labor with AI unfold incredibly well for the populations of rich countries, they might dramatically deepen inequality between countries. For the rest of the 21st century, lower-income countries will continue to have young, growing populations in need not of labor-­replacing tech, but of gainful employment. The problem is that machines invented to fill in for missing workers in countries with labor shortages often quickly spread even to countries where unemployment is in the double digits and the majority of the working population is employed by unregistered informal businesses. That is how we find self-service kiosks in South African restaurants and Indian airports, replacing formal-­sector jobs in these and many more countries struggling to create enough of them.  In such a world, many beneficial applications of AI could remain relatively underdeveloped compared with the merely labor-saving ones. For example, efforts to develop AI for climate-­change resilience, early prediction of natural disasters, or affordable personalized tutoring might end up taking a back seat to projects geared to cutting labor costs in retail, hospitality, and transportation. Deliberate, large-scale efforts by governments, development banks, and philanthropies will be needed to make sure AI is used to help address the needs of poorer countries, not only richer ones. The budgets for such efforts are currently quite small, leaving AI on its default path—which is far from inclusive.  But default is not destiny. We could choose to channel more public R&D efforts toward pressing global challenges like accelerating the green transition and improving educational outcomes. We could invest more in creating and supporting AI development hubs in lower-income countries. Policy choices that allow for greater labor mobility would help create a more balanced distribution of the working-age population between countries and relieve the economic pressures that would drive commercial AI to displace jobs. If we do none of that, distorted incentives will continue to shape this powerful technology, leading to profound negative consequences not only for lower-income countries but for everyone.  Artificial intelligence Here’s how people are actually using AI Something peculiar and slightly unexpected has happened: people have started forming relationships with AI systems. Melissa Heikkilä archive page Google DeepMind trained a robot to beat humans at table tennis It was able to draw on vast amounts of data to refine its playing style and adjust its tactics as matches progressed. Rhiannon Williams archive page We need to prepare for ‘addictive intelligence’ The allure of AI companions is hard to resist. Here’s how innovation in regulation can help protect people. Robert Mahari archive page Pat Pataranutaporn archive page A new public database lists all the ways AI could go wrong Its creators hope their work could lead to further research to determine which risks to take more seriously. Scott J Mulligan archive page Stay connected Get the latest updates from mit technology review. Discover special offers, top stories, upcoming events, and more. Thank you for submitting your email! It looks like something went wrong. We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at [email protected] with a list of newsletters you’d like to receive. US stocks are overvalued because of unrealistic expectations for AI-powered economic growth, Vanguard says Investors are too optimistic about the near-term prospects of AI, Vanguard said. Firms would need to growth profit by 40% annually for the next three years to match valuations, the firm said. "This is double the annualized rate of the 1920s, when electricity lit up the nation," Vanguard wrote. With tech companies still pushing the boundaries of artificial intelligence, market excitement for it seems endless. But this enthusiasm expects too much from the technology in too little time, Vanguard wrote on Thursday. Wall Street is rife with upbeat forecasts about what AI could do to the economy and corporate profits. Most of them are pinned to a US workplace revolution and a productivity boom. That optimism has helped fuel strong stock gains, with the benchmark S&P 500 up 18% year-to-date through Thursday. But Vanguard global chief economist Joe Davis thinks expectations are too high, and says that stocks are overvalued even if the AI boom plays out as anticipated. He estimates that US corporate profits would have to growth by 40% annually over the next three years to justify where stocks are trading now. For context, the S&P 500's trailing one-year earnings growth rate through the second quarter of 2024 was 10.9%, according to FactSet data . "I'm optimistic about the long-term potential of artificial intelligence to power big increases in worker productivity and economic growth," global chief economist Joe Davis wrote. "But I'm pessimistic that AI can justify lofty equity valuations or save us from an economic soft patch this year or next." He continued: "This is double the annualized rate of the 1920s, when electricity lit up the nation — not to mention economic output and corporate income statements." Such a historic surge in corporate performance looks even less probable if the economy cools down next year. Vanguard expects GDP to expand by just 1% to 1.5% in 2025. It's not that the investment firm has no faith in AI's potential — its research suggests 45% to 55% odds that AI will trigger a boom in labor productivity. Between 2028 and 2040, that could spur a 3.1% annualized rate of US growth in real terms. But investors need to let go of any notions that this will happen immediately, Davis said. While firms have poured billions to advance their position in the sector, some market players are incorrect in thinking that AI investing will reach $1 trillion in the near term: "$1 trillion in AI investment by 2025 would require 286% growth. That's probably not going to happen, which means we're unlikely to experience an AI-driven economic boom in 2025," he said. Some on Wall Street are much more pessimistic. BlackRock has said there's a strong chance that heavy AI investing will trigger higher inflation before any production boom can come. That could erode corporate profit growth. Main content We've detected unusual activity from your computer network To continue, please click the box below to let us know you're not a robot. Why did this happen? Please make sure your browser supports JavaScript and cookies and that you are not blocking them from loading. For more information you can review our Terms of Service and Cookie Policy . For inquiries related to this message please contact our support team and provide the reference ID below. Artificial Intelligence and Economic Growth This paper examines the potential impact of artificial intelligence (AI) on economic growth. We model AI as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but AI now seems poised to automate many tasks once thought to be out of reach, from driving cars to making medical recommendations and beyond. How will this affect economic growth and the division of income between labor and capital? What about the potential emergence of “singularities” and “superintelligence," concepts that animate many discussions in the machine-intelligence community? How will the linkages between AI and growth be mediated by firm-level considerations, including organization and market structure? The goal throughout is to refine a set of critical questions about AI and economic growth, and to contribute to shaping an agenda for the field. One theme that emerges is based on Baumol’s “cost disease” insight: growth may be constrained not by what we are good at but rather by what is essential and yet hard to improve. MARC RIS BibTeΧ Download Citation Data Comment on "Artificial Intelligence and Economic Growth" Published From Paper More from nber. In addition to working papers , the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter , the NBER Digest , the Bulletin on Retirement and Disability , the Bulletin on Health , and the Bulletin on Entrepreneurship  — as well as online conference reports , video lectures , and interviews . Try AI-powered search Europe’s economic growth is extremely fragile Risk is concentrated in one country: Germany Your browser does not support the <audio> element. W hen an economy contracts for two consecutive quarters, it is often considered to be in recession. European policymakers will be hoping that two consecutive quarters of growth are equally notable. Data released on August 14th showed that, in the second quarter of the year, the EU ’s economy once again grew by 0.3% against the previous quarter. Although nothing to write home about by American standards, such growth is a relief after more than a year of stagnation. The good news does not stop there. Employment is growing, albeit more slowly than before. Wage growth is outpacing inflation, too, leading to rising living standards. In the Netherlands, which has the continent’s most up-to-date labour-market data, centrally negotiated wages rose by 7% in July, twice the pace of inflation. Union-negotiated wages are similarly strong in Germany. Nevertheless, the European Central Bank ( ECB ) still felt confident enough to cut interest rates in June, and is expected to do so again in September. Full speed ahead, then? Not quite. The continent faces a number of risks—any of which could make the picture much gloomier. The first is that demand does not look as healthy as growth figures might suggest, as is illustrated by the construction industry. Rents are rising in many of Europe’s most alluring cities: Athens, Berlin and Madrid are all seeing growth of about 10% a year. On top of this, interest rates are falling, which should boost property prices. Yet housebuilder confidence is now at the lowest it has been this year, for reasons that are not immediately clear. Growth in incomes should also be boosting consumption. In reality, however, “we are yet to see any meaningful pickup in real domestic demand,” observes Clemente De Lucia of Deutsche Bank. Households are mostly putting the additional money from higher pay into their savings accounts, he adds. In time, a cooling labour market could further reduce the desire to spend. As Davide Oneglia of TS Lombard, a consultancy, notes, hiring has weakened in services, which has been the main source of jobs in recent years. Governments are unlikely to support demand with extra spending of their own. Germany’s has once again almost torn itself apart over the legal intricacies of its balanced-budget rules. Negotiations are ongoing, but the result is likely to be spending cuts. France and Italy, meanwhile, are both in an “excessive deficit procedure”, which the European Commission reserves for the most blatant violators of its guidelines. As such, fiscal policy will be a drag on growth in the years to come. The next worry concerns a single country: Germany. It has barely grown since 2019. More recently, its exports fell by 4.4% in June on a nominal basis, compared with a year earlier, and surveys indicate that worse is to come. Industrial companies that have failed to modernise now face a bigger challenge from China, as low-cost electric vehicles ( EV s) pour out of its factories. Germany’s long-term prospects are also concerning: other than Lithuania, no country in the OECD is set to lose more workers to retirement, relative to new entrants into the labour force. The country is big enough that its economic woes will also drag on Europe’s growth. The continent’s trading partners will not come to the rescue. American demand, though enviable, is starting to weaken and China’s economy is in a mess, which officials are hoping to fix with manufacturing subsidies. If Donald Trump is elected, trade wars—both transatlantic and between America and China—will worsen the situation. Europe’s conflict with China is already under way, as the country prepares to sue the EU at the World Trade Organisation for raising tariffs on EV s. As it stands, Europe appears to be pulling off a soft landing, even if its economy never truly soared in the first place. Inflation has fallen to 2.5%, just above the ECB ’s target, and the continent has enjoyed two consecutive quarters of growth. But the euro zone’s policymakers would be wise not to take too much cheer from this. Plenty of dangers must first be navigated before the celebrations can begin. ■ For more expert analysis of the biggest stories in economics, finance and markets, sign up to  Money Talks , our weekly subscriber-only newsletter. Explore more This article appeared in the Finance & economics section of the print edition under the headline “Top of the danger list” Finance & economics August 17th 2024 Vladimir Putin spends big—and sends Russia’s economy soaring How vulnerable is Israel to sanctions? How to invest in chaotic markets What is behind China’s perplexing bond-market intervention? Why companies get inflation wrong From the August 17th 2024 edition Discover stories from this section and more in the list of contents More from Finance & economics How Vladimir Putin hopes to transform Russian trade He believes the country’s future lies with China and India. What could go wrong? Vast government debts are riskier than they appear A provocative new paper gets central bankers talking at Jackson Hole Jerome Powell (almost) declares victory over inflation The Federal Reserve chairman strikes a notably doveish tone Investors should avoid a new generation of rip-off ETFs Some proposals may even be a risk to financial stability Why investors are not buying Europe’s revival Even though the continent’s stocks are in a “sweet spot” America’s recession signals are flashing red. Don’t believe them We assess a range of measures We use cookies to ensure best experience for you We use cookies and other tracking technologies to improve your browsing experience on our site, show personalize content and targeted ads, analyze site traffic, and understand where our audience is coming from. You can also read our privacy policy , We use cookies to ensure the best experience for you on our website. Leaders Speak Apple’s global revenues to cross $400 bn mark in 2024, AI to boost services segment growth: Counterpoint Updated On Aug 28, 2024 at 11:15 AM IST Published On Aug 28, 2024 at 11:15 AM IST All Comments By commenting, you agree to the Prohibited Content Policy Find this Comment Offensive? Foul Language Inciting hatred against a certain community Out of Context / Spam Join the community of 2M+ industry professionals Subscribe to our newsletter to get latest insights & analysis., download ettelecom app. Get Realtime updates Save your favourite articles telecom news technology news Counterpoint apple intelligence 210 IMAGES Artificial Intelligence Economy (PDF) The Economics of Artificial Intelligence Science and Technology as the Engine of Economic Growth and Essay SOLUTION: Artificial intelligence and economic growth (PDF) Artificial Intelligence and Economic Growth The Future Of Growth And Economic Development Powered By Human And COMMENTS Artificial Intelligence and Economic Growth DOI 10.3386/w23928. Issue Date October 2017. This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century ... PDF Artificial Intelligence and Economic Growth Artificial Intelligence and Economic Growth. 237. 9.1 Introduction. This chapter considers the implications of artifi cial intelligence for eco- nomic growth. Artifi cial intelligence (AI) can be defi ned as "the capability of a machine to imitate intelligent human behavior" or "an agent's ability to achieve goals in a wide range of ... PDF AI and the Economy We review the evidence that artificial intelligence (AI) is having a large effect on the economy. ... Slowing economic growth over the past decade underscores the importance of AI to deliver on its potential productivity benefits. Furman (2017) reports that 36 of 37 advanced ... academic papers focused on AI have increased 9 times since 1996 ... Artificial Intelligence and Economic Development: An Evolutionary In today's environment of the rapid rise of artificial intelligence (AI), debate continues about whether it has beneficial effects on economic development. However, there is only a fragmented perception of what role and place AI technology actually plays in economic development (ED). In this paper, we pioneer the research by focusing our detective work and discussion on the intersection of ... PDF Artificial Intelligence and Economic Growth Abstract. f artif. cial intelligence (A.I.) on economicgrowth. We start by modeling A.I. as. this. erspective in light of the evidence todate. We further discuss linkages betw. en A.I. and growth as mediated by firm-levelconsiderations, including organiza. perintelligence" that animate many discus-. sions in the machine intelligence comm. PDF Artificial Intelligence and Economic Growth A.I. modeled as a continuation of automation. Automation = replace labor in particular tasks with machines and algorithms. ms, steam engines, electric power, computers Future: driverless cars, paralegals, pat. A.I. may be limited by Baumol's cost disease. Baumol: growth constrained not by what we do well but rather by what is essential and ... PDF Artificial Intelligence and Economic Growth artificial intelligence for economic growth. Artificial intelligence (A.I.) can be defined as "the capability of a machine to imitate intelligent human behavior" or "an agent's ability to achieve goals in a wide range of environments."1 These definitions i. Implications of AI innovation on economic growth: a panel data study The application of artificial intelligence (AI) across firms and industries warrants a line of research focused on determining its overall effect on economic variables. As a general-purpose technology (GPT), for example, AI helps in the production, marketing, and customer acquisition of firms, increasing their productivity and consumer reach. Aside from these, other effects of AI include ... PDF Artificial Intelligence and Economic Growth artificial intelligence for economic growth. Artificial intelligence (A.I.) can be defined as "the capability of a machine to imitate intelligent human behavior" or "an agent's ability to achieve goals in a wide range of environments."1 These definitions i. Artificial Intelligence and Economic Growth Artificial Intelligence and Economic Growth. Philippe Aghion, Benjamin Jones and Charles Jones. No 23928, NBER Working Papers from National Bureau of Economic Research, Inc Abstract: This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Artificial Intelligence, Globalization, and Strategies for Economic DOI 10.3386/w28453. Issue Date February 2021. Progress in artificial intelligence and related forms of automation technologies threatens to reverse the gains that developing countries and emerging markets have experienced from integrating into the world economy over the past half century, aggravating poverty and inequality. The new technologies ... (PDF) Artificial Intelligence and Economic Growth: A Theoretical Scientific Annals of Economics and Business 68 (4), 2021, 421-443 DOI: 10.47743/saeb-2021-0027 Artificial Intelligence and Economic Growth: A Theoretical Framework Lei Wang*, Provash Kumer Sarker** , Kausar Alam*** , Shahneoaj Sumon § * Abstract The growing adoption of Artificial Intelligence (AI) has sparked ubiquitous concerns worldwide. Artificial Intelligence and Economic Growth This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process ... Benjamin F. and Jones, Charles I., Artificial Intelligence and Economic Growth (October 2017). NBER Working Paper No. w23928, Available at SSRN: https ... PAPERS. 7,432. Political ... OECD Artificial Intelligence Papers The impact of Artificial Intelligence on productivity, distribution and growth Key mechanisms, initial evidence and policy challenges This paper explores the economics of Artificial Intelligence (AI), focusing on its potential as a new General-Purpose Technology that can significantly influence economic productivity and societal wellbeing. Artificial Intelligence and Economic Growth The potential impact of artificial intelligence on economic growth is examined, based on Baumol's "cost disease" insight: growth may be constrained not by what the authors are good at but rather by what is essential and yet hard to improve. This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a ... 9 Artificial Intelligence and Economic Growth Abstract. This chapter examines the potential impact of artificial intelligence (AI) on economic growth. We model AI as the latest form of automation, a broader process dating back more than 200 years. Artificial Intelligence and Economic Growth Working Papers; Filter Publications by Topic; ... Artificial Intelligence and Economic Growth. This paper examines the potential impact of artificial intelligence (A.I.) on economic growth. We model A.I. as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and ... The Impact of Artificial Intelligence on Economic Growth From the Artificial intelligence has sophisticated social and economic effects that cannot be ignored. Based on a thorough review of the development of artificial intelligence, this paper systematically explores the mechanism of the impact of artificial intelligence on economic growth through technology, value and application three paths, which is starting from the perspective of the population ... The impact of artificial intelligence on economic growth and welfare Introduction. The purpose of this paper is to construct a basic endogenous growth model with artificial intelligence (AI), and to explore the impact of the development of AI on economic growth and welfare. The innovation of this paper is that we incorporate the self-accumulation ability and the nonrival characteristic of AI into the model. AI and the Economy: Innovation Policy and the Economy: Vol 19 Executive Summary We review the evidence that artificial intelligence (AI) is having a large effect on the economy. Across a variety of statistics—including robotics shipments, AI start-ups, and patent counts—there is evidence of a large increase in AI-related activity. We also review recent research in this area that suggests that AI and robotics have the potential to increase ... The Macroeconomics of Artificial Intelligence AI may affect society in a number of areas besides the economy—including national security, politics, and culture. But in this article, we focus on the implications of AI on three broad areas of macroeconomic interest: productivity growth, the labor market, and industrial concentration. AI does not have a predetermined future. 9. Artificial Intelligence and Economic Growth Aghion, Philippe, Jones, Benjamin F. and Jones, Charles I.. "9. Artificial Intelligence and Economic Growth" In The Economics of Artificial Intelligence: An Agenda edited by Ajay Agrawal, Joshua Gans and Avi Goldfarb, 237-290. Chicago: University of Chicago Press, 2019. Will the AI Revolution Lead to Greater Prosperity? CAMBRIDGE - As global economic growth slows, many hope technological innovation is a potential solution. The International Monetary Fund's latest World Economic Outlook, for example, highlighted the potential of artificial intelligence to boost productivity and GDP.But the report also warns that given the uncertainties surrounding the extent of AI's impact, such forecasts should be ... Maximizing Economic Impact: How to Fine-Tune AI Artificial intelligence has grown to become a force in the global economy; its wide possibility is for unprecedented growth, efficiency, and innovation. The optimization of the described AI systems for maximum economic impact must be about adjusting the AI technologies to ensure effective productivity and growth and address some of humanity's ... What will AI mean for economic inequality? If we're not careful, we could see widening gaps within countries and between them. Prominent AI researchers expect the arrival of artificial general intelligence anywhere between "the next ... Stock Market Overvalued Vs. the Profit Growth AI Will Deliver: Vanguard Firms would need to growth profit by 40% annually for the next three years to match valuations, the firm said. "This is double the annualized rate of the 1920s, when electricity lit up the nation ... South Korea's Early Exports Pick Up on Resilient Chip Demand South Korea's early trade data showed exports gaining momentum this month, boosting the outlook for economic growth and suggesting global demand for semiconductors remains resilient. Artificial Intelligence and Economic Growth Artificial Intelligence and Economic Growth. This paper examines the potential impact of artificial intelligence (AI) on economic growth. We model AI as the latest form of automation, a broader process dating back more than 200 years. Electricity, internal combustion engines, and semiconductors facilitated automation in the last century, but AI ... Europe's economic growth is extremely fragile Wage growth is outpacing inflation, too, leading to rising living standards. In the Netherlands, which has the continent's most up-to-date labour-market data, centrally negotiated wages rose by ... Apple's global revenues to cross $400 bn mark in 2024, AI to boost Apple will hold its fall event at its US headquarters on September 9, where it is likely to launch its new iPhone range, and updates to other devices and services, with artificial intelligence (AI ... assignment essay homework paper phd report resume review thesis