Government Artificial Intelligence readiness and brain drain: Influencing factors and spatial effects in the European Union member states

Iuga, Iulia Cristina; Socol, Adela Article Government Artificial Intelligence readiness and brain drain: Influencing factors and spatial effects in the European Union member states Journal of Business Economics and Management (JBEM) Provided in Cooperation with: Vilnius Gediminas Technical University (VILNIUS TECH) Suggested Citation: Iuga, Iulia Cristina; Socol, Adela (2024) : Government Artificial Intelligence readiness and brain drain: Influencing factors and spatial effects in the European Union member states, Journal of Business Economics and Management (JBEM), ISSN 2029-4433, Vilnius Gediminas Technical University, Vilnius, Vol. 25, Iss. 2, pp. 268-296, https://doi.org/10.3846/jbem.2024.21136 This Version is available at: https://hdl.handle.net/10419/317679 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright © 2024 The Author(s). Published by Vilnius Gediminas Technical University ISSN 1611-1699 / eISSN 2029-4433 2024 Volume 25 Issue 2 Pages 268–296 https://doi.org/10.3846/jbem.2024.21136 GOVERNMENT ARTIFICIAL INTELLIGENCE READINESS AND BRAIN DRAIN: INFLUENCING FACTORS AND SPATIAL EFFECTS IN THE EUROPEAN UNION MEMBER STATES Iulia Cristina IUGA , Adela SOCOL Department of Finance-Accounting, Faculty of Economic Sciences, “1 Decembrie 1918” University of Alba Iulia, Alba Iulia, Romania Article History: Abstract. In the swiftly advancing field of Artificial Intelligence (AI), a field where every country aims to keep pace, significant disparities are observed in how different nations adopt AI. This study explores the deep, yet insufficiently studied, effects of AI on societal, economic, and environmental aspects. It particularly examines how brain drain influences governmental AI implementation capabilities, addressing a gap in existing literature. The study investigates the interplay between government AI implementation and brain drain, factoring in macroeconomic conditions, governance quality, educational levels, and R&D efforts. Utilizing 2022 data from European Union countries, the research employs instrumental-variables regressions (2SLS and LIML) to counteract endogeneity and uses clustering methods for categorizing countries based on their government AI levels, alongside spatial analysis to detect cross-national spillovers and interactions. The findings reveal brain drain’s detrimental effect on governmental AI preparedness, highlight clustering tendencies, and identify spatial interdependencies. This paper underscores the need for strategic policy-making and institutional reforms to bolster government AI capabilities. It advocates for a paradigm shift in government frameworks post-New Public Management era, tailored to the new challenges posed by AI. The research, however, is limited to a single year and region, with constraints on data availability and indicator breadth. ■ received 29 May 2023 ■ accepted 1 February 2024 Keywords: government AI readiness, brain drain, government expenditure, spatial effects, spillover effects, human capital. JEL Classification: O15, H52, O38. Corresponding author. E-mail: [email protected] JOURNAL of BUSINESS ECONOMICS & MANAGEMENT 1. Introduction The concept of Artificial Intelligence (AI), often utilized in specialized studies and increasingly prevalent in recent language, encompasses technically complex and computer-oriented ideas that are difficult to succinctly express. From a governmental standpoint, AI represents a multifaceted blend of technology, policy, and social impact, necessitating a multidisciplinary approach for its successful integration and application. In citizen services, AI’s applications range from handling inquiries and processing documents to directing requests, aiding translations, and drafting documents (Mehr, 2017). Three primary AI applications in government stand out: Robotic and cognitive automation, enabling the reallocation of human labor to more value-added tasks through technologies like Robotic Journal of Business Economics and Management, 2024, 25(2), 268–296 269 Process Automation; Cognitive insights generation, improving predictive capacities; and Cognitive engagement, addressing citizen queries (Eggers et al., 2017). Government entities are increasingly adopting AI, expected to fundamentally transform their operations, from policy-making to the execution of daily public services. AI provides tools for predictive analytics, decision-making, and problem-solving, particularly valuable in contexts of significant uncertainty. Utilizing AI capabilities allows public sector organizations to enhance agility, anticipate citizen needs, optimize resource allocation, and improve service delivery quality (Mikhaylov et al., 2018). AI presents countries with a vital opportunity to advance their economic landscapes, especially in public services. It enables enhanced, personalized service delivery (Robles & Mallinson, 2023). Its powerful data analytics capacity helps governments derive insights from large data sets, supporting evidence-based policymaking. AI equips governments with powerful tools for automating bureaucratic tasks, extracting insights from vast data, and customizing public services to meet individual needs, thus symbolizing a beacon for improved governance, informed policymaking, and heightened citizen engagement (Frățilă et al., 2023). The European Union, a unique assembly of countries with varied economic and technological paths, offers a key study area for examining government AI readiness and brain drain within its unified political and economic structure. Understanding each member state’s AI preparedness is crucial, as disparities could significantly impact the EU’s collective policy and strategic direction. The EU’s principle of free movement (Article 45, Treaty on the Functioning of the EU (European Union, 2012)) highlights its unique brain drain issues, with high-skilled workers easily relocating between states, differing from global patterns. Member states face similar problems like aging populations (Walker & Maltby, 2012) and upskilling needs (Jacobs, 2023). Examining AI readiness in this context reveals opportunities for collective strategies and cooperation. Given the EU’s acknowledgment of AI as an instrumental force shaping future socioeconomic contours (European Parliament, 2023), gauging the AI readiness of its governments is indispensable. “The Government AI Readiness Index provides an overall estimate for how prepared each country’s national government is for implementing AI in public service delivery” (Oxford Insights, 2022). The EU’s policy harmonization, alongside its member nations’ varied economic and technological stages (Calderaro & Blumfelde, 2022), poses challenges and opportunities in tackling brain drain and enhancing AI readiness. The EU’s Coordinated Plan on Artificial Intelligence (European Commission, 2018) encouraged nations to develop or integrate AI strategies. By 2023, all 27 EU countries had AI strategies (OECD, 2023; European Commission, 2020; Cath et al., 2017). Studying the EU offers a comprehensive view on AI readiness and brain drain dynamics, yielding relevant insights into AI integration and labor mobility trends in the EU. Prior research reveals new emerging paradigms in public administration (post-New Public Management Paradigms) have been identified that focus on the importance of technological innovation and artificial intelligence to improve the delivery of services to citizens and the policies implemented by the government (Ojo et al., 2019). These new paradigms focus on tangible benefits and creating public value through investment in disruptive technologies. Integrating AI into governance introduces innovation but also challenges. Studies indicate that governments struggle to keep pace with AI advancements, showing a noticeable gap in public sector AI governance (Wirtz et al., 2020). Challenges often arise from skill shortages 270 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... (Dwivedi et al., 2021), evident in employees’ limited AI and machine learning knowledge (Ojo et al., 2019) and a lack of local AI experts (Gupta, 2019). The scarcity of AI professionals (Al-Mushayt, 2019) and insufficient advanced technology education (Montoya & Rivas, 2019) add to these difficulties. Simultaneously, many countries, especially developing ones, face a significant “Brain drain” – skilled workers migrating to developed countries for better opportunities. The Fragile States Index highlights the economic and developmental impacts of such human displacement (Fund for Peace, 2022). In the context of the European Union (EU), spatial interactions and spillovers between member states are crucial for understanding phenomena like brain drain and AI readiness. In the EU, countries are interdependent; developments in one can influence others. This interconnectedness, inherent to the EU’s structure, necessitates a comprehensive view of these dynamics, considering the collective impact rather than isolated national trends. The EU’s principle of free movement for workers fosters a unified labor market, enabling professionals to pursue opportunities across borders with fewer migration barriers. This mobility carries significant consequences: professionals transfer their skills, experiences, and insights as they move (Frankowska & Pawlik, 2022), encouraging cross-border collaborations in research and innovation. The interconnectedness of the EU means that member states are continually observing, learning from, and adapting policies from their neighbors (Altuzarra et al., 2019). A successful policy or initiative in one country can serve as a blueprint for others. To our knowledge, there has been no research exploring within the EU member states how the national context influences government readiness to adopt AI in neighboring states. The aim of the study is to analyze whether governments’ readiness for artificial intelligence AI in the European Union countries depends on “brain drain”, using Government Artificial Intelligence Readiness Index as dependent variable and the Human Flight and brain drain as the major independent proxy. Also, given the specifics of the EU labour market and the freedom of movement of workers, we were interested in discovering possible spatial interactions and spillover effects between countries. The objective of the paper is threefold: 1. Examines the influence of brain drain on government AI readiness in EU countries; 2. Studies the existence of spatial effects and spillover effects between neighboring EU countries; 3. Highlight the policies that can be adopted to reduce the government AI discrepancy between the EU member states. This study addresses the following research questions: ■RQ1: Does the European Union experience a negative effect of brain drain on AI government readiness? ■RQ2: Are there positive effects of EU-funded projects for R&D, economic freedom, and government spending on AI government readiness in the European Union countries? ■RQ3: Does the brain drain (along with control variables) in a country influences the government’s readiness to adopt AI in the neighboring states? While there is indirect literature exploring the individual phenomena of brain drain and government AI readiness, there remains a paucity of research specifically examining the interplay between these two dimensions, especially within the unique context of the European Union. The existing body of work often treats brain drain and AI readiness as distinct entities, with limited exploration of how they might influence each other. Moreover, the spatial interactions and spillover effects within the EU, shaped by its singular policies such as the Journal of Business Economics and Management, 2024, 25(2), 268–296 271 freedom of movement, present a nuanced backdrop that hasn’t been adequately addressed in current studies. This lacuna in understanding becomes particularly significant given the rapid technological advancements and shifting labor dynamics within the EU. As such, there is a pressing need for an integrative analysis that holistically examines how brain drain might impact AI readiness across governments and vice versa, and how these dynamics manifest in the EU’s interconnected landscape. This study endeavors to bridge this gap, offering insights that can inform both policy-making and future academic pursuits. The present study bridges gaps in the prevailing literature on AI in government and the phenomenon of brain drain. Historically, AI’s role in public sector augmentation has been discussed in isolation, with emphasis on its potential to streamline operations and elevate citizen engagement. This study illuminates the intricate interplay between brain drain and government AI readiness, a dimension underexplored in earlier works. Echoing findings from Oxford Insights (2019) we underscore the criticality of human capital in the AI realm. This aligns with Docquier and Rapoport (2012), who highlighted the repercussions of high-skilled labor emigration on a country’s technical prowess. Our research deviates from merely acknowledging brain drain as a challenge, extending into its direct and indirect impacts on AI adoption, especially in the European context. Furthermore, while past literature like Brynjolfsson and McAfee (2014) touched upon potential delays in AI development due to expertise shortage, our findings delve deeper into the cascading effects of such delays on public service quality and national security. By juxtaposing economic freedom, EU-funded R&D projects, and government readiness for AI, we offer a comprehensive narrative that not only augments existing literature but also charts out potential trajectories for future research in the domain. The innovation of this paper is that we incorporate five major groups of factors (human, macroeconomic, governance, education and research & development) into the model and study both panel effects for European Union states, as well as spatial effects and data clustering. The research’s originality lies in its unique combination of geographical focus, methodological robustness, emphasis on spatial interactions, exploration of the brain drain phenomenon, and its bridge between theoretical findings and policy implications. The remainder of this paper is organized as follows: Section 2 outlines the current state of knowledge in the field and the research hypothesis; Section 3 details the data and methodology; Section 4 focuses on the results and discussion; and the paper concludes with the conclusions. 2. Literature review 2.1. Interaction between government AI and brain drain Government AI signifies the use of AI technologies in the public sector, enhancing efficiency, effectiveness, and decision-making. It possesses the potential to transform government operations by improving citizen services, reducing costs, and optimizing resource use. AI involves computational techniques that enable machines to learn, reason, and solve problems, akin to human cognition. Worldwide, governments are increasingly interested in embedding AI into their operational fabric. AI’s application in government spans various domains: in Public Services, it could automate tasks like application processing and inquiry handling; in Policy and Decision-making, it could analyze extensive datasets for policy shaping; in Law Enforcement and Public Safety, 272 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... technologies like facial recognition and predictive analytics could augment safety efforts; in Infrastructure and Resource Management, AI could assist in resource management and infrastructure planning (Wang & Cui, 2022); and in Healthcare, AI could improve public health outcomes by analyzing medical data for disease detection and treatment optimization (Gomes de Sousa et al., 2019). Countries’ readiness for AI technology varies, as revealed by an Oxford Insights (2019) review, which identified challenges in adopting AI for the common good, including policy, capacity, and resources, with human capital as a key resource. The Human flight and brain drain indicator focuses on the economic effects of skilled labor migration and its impact on a nation’s advancement. Brain drain, a complex phenomenon, particularly affects developing countries, with skilled workers emigrating to more developed regions (Docquier & Rapoport, 2012). This migration impacts a nation’s capacity to maintain and grow its expertise in critical areas like AI. In the context of globalization and international competition, brain drain affects a country’s ability to maintain and develop its expertise in key areas such as AI and attract foreign direct investment (Czaika & de Haas, 2015; Siar, 2013). In the European Union, there’s a notable migration of experts from Central and South-Eastern Europe to Western countries (Bălan & Olteanu, 2017). This movement, driven by personal and socio-economic factors, began after the fall of communist regimes in the 1990s. The impact of brain drain on government AI is multi-dimensional: It restricts access to needed expertise for AI development (Brynjolfsson & McAfee, 2014), potentially causing delays in AI adoption and risks to data protection (Dignum, 2019). It also leads to a loss of AI talent to the private sector, slowing public sector innovation. Additionally, governments might become over-reliant on private AI solutions (Gesk & Leyer, 2022), posing national security risks and loss of control over data and algorithms (Offer, 2022). The public-private pay gap (Agrawal et al., 2019) and the lack of public sector investment in AI R&D (Johnson, 1965) further fuel brain drain. Overall, while AI holds transformative potential for government operations, its effective integration faces hurdles due to brain drain, resource limitations, and evolving AI development and policy landscapes. This interplay requires a multifaceted approach to enhance AI readiness in the public sector, considering the dynamics of economic disparities and global talent mobility. Considering these, we can formulate the main research hypothesis of this study: H1: There is a negative effect of brain drain on AI government readiness in the European Union countries. 2.2. Interaction between macroeconomic, governance, education and research & development variables on AI government readiness Brain drain, government spending, economic freedom and EU-funded research and development projects for artificial intelligence are fundamental themes for the contemporary debate on sustainable development and national or European competitiveness. The introduction of artificial intelligence into government operations has the potential to transform the mode the public sector operates, optimizing efficiency, effectiveness and decision-making. Government spending can have a positive impact on government readiness for AI, ensuring that public institutions can reap the benefits of this innovative technology (Bredt, 2019). Journal of Business Economics and Management, 2024, 25(2), 268–296 273 Investing in infrastructure, health, education and research can help increase a country’s competitiveness and improve citizens’ quality of life (Bose et al., 2007), while the development of the telecommunication infrastructure has a significant positive impact on the efficiency of government (Doran et al., 2023). Thus, governments can also help prepare the workforce for the integration of AI in the public sector. By investing in training and reskilling programs, governments can ensure that public sector employees are prepared to work with AI technologies and manage the changes associated with the automation of certain tasks (Duan et al., 2019). This can lead to a better adaptation of the workforce to technological developments and to a smoother transition into the digital age. With the development of advanced language models using Artificial Intelligence (AI), the issue of AI readiness becomes critical, and the quality of the human factor involved in AI depends largely on the use of AI for purposes truly useful to humanity and without associated risks, especially those related to unpredictability and ethical concerns. Another important aspect of government spending is investment in the infrastructure needed to support AI implementation (Wang & Cui, 2022). This may include the development of high-speed communication networks, data centers and other technological resources that enable the efficient use of AI in public services and decision-making processes. On the other hand, investments in infrastructure, education and research and development can help increase a country’s AI competitiveness. Studies show that governments investing in AI can benefit from economic growth and improved quality of life for citizens (Arntz et al., 2019). Government investment in AI can also foster international cooperation (Millard, 2017) and partnerships between different governments and organizations, promoting a global and harmonized approach to AI regulations and standards. This can lead to greater interoperability between AI systems used in different countries and to strengthen collaboration to address common challenges such as cybersecurity and data protection (Pan & Zhang, 2021). Government spending on R&D in AI is another key issue for sustainable economic development. The European Commission has made significant investments in the projects of Artificial Intelligence (AI) research and development firms. This is done with the intention of enhancing the preparedness of governments to adopt AI technologies. Given the premise that governmental expenditures are directed towards enhancing the capability of governments to deploy artificial intelligence, it is hypothesized that: H2: There is a positive effect of government spending on AI government readiness in the European Union countries. The Economic Freedom Index by The Heritage Foundation assesses economic liberty worldwide, often used to examine the economic conditions favorable for Artificial Intelligence (AI) growth. Nations scoring higher on this index typically present a conducive environment for AI adoption, characterized by lower taxes, fewer regulations, and better property rights protection. Economic freedom is instrumental for governments in AI readiness, mainly by attracting investment and fostering innovation. Countries with high scores in the Index of Economic Freedom attract investors, thanks to stable economic conditions and growth potential. This environment helps governments gather necessary capital and expertise for AI development and implementation (World Economic Forum, 2023). Moreover, countries with greater economic freedom usually have dynamic, competitive markets, sparking innovation and advanced AI technologies. In such markets, companies invest more in research and development, striving to create and market top-tier AI products (Ciftci & Durusu-Ciftci, 2022). 274 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... Another benefit of economic freedom for government AI readiness is the promotion of entrepreneurial spirit and risk-taking. High-scoring countries on the Index of Economic Freedom offer favorable conditions for entrepreneurship, like easier credit access and strong property rights. These factors encourage individuals and businesses to pursue innovative ideas, including AI-related ventures. As a result, governments in these countries have access to a continuous stream of innovative AI technologies and applications (Le & Kim, 2020). Economic freedom also facilitates the integration of AI technologies within governments. High-ranking countries on the Index of Economic Freedom typically have efficient, streamlined bureaucratic systems that can easily adapt to AI-driven changes. These nations face fewer bureaucratic hurdles to innovation, allowing government entities to experiment with new AI applications and technologies more freely (Okulich-Kazarin et al., 2020). In summary, economic freedom significantly influences government AI readiness. It draws investment, stimulates innovation, fosters entrepreneurship, and eases AI technology integration within government. Countries prioritizing economic liberty are better positioned to leverage AI’s benefits, thus reaping the rewards of this transformative technology. Considering these, we can estate: H3: There is a positive effect of economic freedom on AI government readiness in the European Union countries. EU-funded projects for R&D in AI are another key element in supporting technological development and innovation. The European Union has invested massively in programs such as Horizon 2020 and the new Horizon Europe program (2021–2027) to boost AI research and development (European Commission, 2021a). These initiatives offer financial support and collaborative prospects among researchers, universities, and companies across various European countries, thereby advancing the sharing of knowledge and the creation of innovative solutions (Spence, 2021). A pertinent example is the AI4EU project, funded by the EU under Horizon 2020. It strives to establish a European platform for the advancement and utilization of AI, contributing positively to both the economy and society (AIoD Platform, 2019). The substantial investment by the European Commission in projects related to AI research and development firms has notably enhanced government readiness for AI. These initiatives have contributed to better government efficiency, increased citizen engagement and satisfaction, and augmented accountability and transparency. As AI technology continues to evolve, these advantages are expected to become even more distinctive, positioning AI as an increasingly vital instrument for governments globally. These projects have produced considerable positive effects, furnishing governments with the essential tools to leverage the potential of AI and revolutionize their operations. A key advantage of the European Commission’s funding in AI research and development firms’ projects lies in the improvement of government operations’ efficiency and efficacy. Utilizing AI-powered instruments, governments can automate mundane tasks and procedures, consequently cutting operational expenses, enhancing accuracy, and boosting speed. For example, AI can assist governments in swiftly and precisely processing enormous quantities of data, facilitating superior decision-making and more efficient resource allocation (European Parliament, 2021a). Another significant benefit of these projects is improved citizen engagement and satisfaction. AI-driven chatbots and virtual assistants can facilitate citizens in accessing government services with greater speed and efficiency, leading to reduced waiting times and an enhanced overall experience. Additionally, AI-enabled tools can assist governments in more Journal of Business Economics and Management, 2024, 25(2), 268–296 275 comprehensively comprehending the needs and preferences of their citizens (Ojo, 2019), thereby allowing for the delivery of more tailored services (European Parliament, 2021b). The investments made by the European Commission in AI research and development firms’ projects have additionally contributed to enhancing government accountability and transparency. AI-enabled tools can aid governments in overseeing and assessing their performance, simplifying the process of pinpointing areas that need improvement and tracking progress over time. Moreover, AI can assist governments in identifying and averting fraud and corruption, thereby guaranteeing that public funds are utilized properly (European Commission, 2021b). Considering the abovementioned, we can formulate the last hypothesis: H4: There is a positive effect of EU-funded projects for R&D on AI government readiness in the European Union countries. 3. Data and methodology Our study analyzes whether governments’ readiness for artificial intelligence AI in the European Union (EU) countries depends on “brain drain”, using Government Artificial Intelligence Readiness Index as dependent variable and the Human flight and brain drain as the major independent proxy. Several economic, research and development and governance factors are considered control variables, based on the results identified in the previous studied literature. To control the macroeconomic conditions, the study use Government Expenditure, Gross Domestic Product and Economic Freedom, while to control governance and education in AI field, two proxies are employed: Government Pillar of AI Index and AI in University Bachelor’s Programs (proportion of programs with AI content in the total number of programs). Also, the study considers the variable represented by AI R&D Firms’ Projects founded by European Commission (percentage of the total number of AI R&D players financed) as a proxy to capture Research and Development activities in AI. The panel comprises the European Union countries, except for Cyprus, Malta and Ireland, which are excluded from the studied sample, given their geographical status without EU neighboring countries (bases on spatial analysis methods that do not allow the analysis of those states that do not have common borders with the rest of the states in the chosen sample). Table 1 presents the description of the variables and data sources from which the information was gathered. To best to our knowledge, at regional or global level, the scarcity of indicators to capture the preparation of governments in the implementation of artificial intelligence is obvious, because there are limited initiatives to develop such indicators. We are interested in capturing the situation of AI implementation in government in the period as close to the present as possible, so we use a year window which refers to the year 2022, and for the rest of the variables we rely on the most recent data collection available, published in 2022 and related to the previous year or years, depending on the collection criteria of those variables. The issue of preparing governments for AI deployment is rarely studied in the literature, much less its connection to brain drain, given the novelty of the subject and the incipient concerns of scientific communities to analyze unconventional perspectives of AI implementation by governments. As far as the author’s knowledge is concerned, no study has explored such a topic from the perspective of the European Union and the variables chosen by us. 282 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... for Anderson-Rubin and not significant for the Basman test). Also, in the 2SLS model, we perform tests to study the explanatory power of the instruments, whose outcomes confirm that additional instruments (lnGDP, GOVERNANCE and AI BACHELOR) have significant explanatory power for AI IN GOVERNANCE after control for the effect of the rest of the independent variables of the model (F statistic significant at 1% level). The minimum eigenvalue statistic as a test of weak instruments compared with critical values of 2SLS relative bias denotes that our instruments are not weak (the F statistic exceeds the critical value if we were willing tolerate a 10% relative bias). As far human proxy is concerned, BRAIN DRAIN has a significant negative influence on AI IN GOVERNMENT, which shows that the increase in the flows of human flight and brain drain from a country that is losing human capital we are witnessing a reduction in the government’s ability to implement and manage artificial intelligence in public services for citizens. There are only circumstantial indications within the cited literature suggesting a negative effect of brain drain on government AI readiness. Our findings are consistent with this body of literature (Czaika & de Haas, 2015; Siar, 2013; Brynjolfsson & McAfee, 2014; Dignum, 2019) and validate the main research hypothesis H1 that brain drain negatively affects AI in government. The brain drain phenomenon can have a negative impact on a country’s government’s AI readiness, for several reasons: 1. Shortage of AI-skilled human resources can slow down or even stop the development and implementation of AI-based solutions in the public sector (Vicsek, 2021); 2. Brain drain can negatively affect the level of innovation in a country (Spence, 2021) (when AI experts leave the country, they take their knowledge and experience with them, which can lead to a decrease in scientific and technological advances in AI); 3. Brain drain can lead to a loss of investment in education and training: governments invest in the education and training of AI specialists, with the hope that they will contribute to the country’s economic development and growth. When these specialists choose to work abroad, the government loses the benefits of its investments, while the host countries benefit from these talents; 4. Brain drain can create major difficulties in international collaboration: experts who leave their home country may be less interested in working with the government and institutions in their country of origin, which can make it difficult to access international knowledge and resource networks; 5. Brain drain can weaken ability to compete globally: as other countries attract talent into AI, they can become more technologically and economically competitive. At the same time, the country of origin could be left behind in the global AI race. Adopting strategies can help governments counter the negative effects of brain drain on government AI training and develop an environment conducive to growing AI expertise and skills in the country. From the multitude of strategies that can be applied by governments, we present the most important ones: 1. Enhancing education systems: Investing in academic and research institutions to deliver superior AI programs, developing AI training and research initiatives, and fostering partnerships between universities, businesses, and governmental entities. Promoting STEM education: Encouraging interest in AI from an early school level through emphasis on science, technology, engineering, and mathematics. 2. Providing financial incentives and support: Allocating government funds for AI research and development and establishing a tax-friendly environment for companies developing and applying AI technologies. Journal of Business Economics and Management, 2024, 25(2), 268–296 283 3. Creating a conducive environment for innovation: Encouraging collaboration between public and private sectors to tackle societal and governmental issues via AI, and facilitating access to essential infrastructure and resources for AI solution development and implementation (Bredt, 2019). 4. Attracting and retaining of AI talent: implementing favorable migration policies to attract and retain foreign AI talent; providing competitive and advantageous career opportunities in the public sector for AI specialists. Even more, the policy of increasing wages and benefits can also be applied (governments can increase wages and benefits for public sector AI professionals to encourage retaining and attracting talent. This strategy could also include the development of bonus schemes and other benefits that would attract and retain AI professionals in the country) (Agrawal et al., 2019). 5. International cooperation and knowledge exchange: participation in international AI alliances, partnerships and projects to benefit from the expertise and resources of other countries; promoting academic and professional exchanges in the field of AI to facilitate the transfer of knowledge and good practice; supporting cooperation between research and innovation institutions in different countries to work together in the development of AI solutions. 6. AI infrastructure development: governments can develop the infrastructure needed to support AI development, such as data centers, communication networks, and cloud computing infrastructure (Duan et al., 2019). This strategy could also include investments in the country’s digital infrastructure, including connectivity and the development of advanced digital technologies. Regarding the control variables of the models, the contribution of GOVERNMENT EXPENDITURE in strengthening the capacity of governments to implement artificial intelligence is confirmed and demonstrates a positive relationship with dependent variable: the higher the volume of government spending, the higher the government’s readiness in artificial intelligence. The results confirm the general assumptions from previous literature (Wang & Cui, 2022; Bredt, 2019; Duan et al., 2019; Bose et al., 2007) and the resorts of such an interdependence are related to the innovative and relatively expensive technical character of the state’s investments in the artificial intelligence infrastructure and software for the public services. Also, the personnel needed to develop and manage informatic applications and infrastructure based on artificial intelligence leads to significant government spending. Given the premise that governmental expenditures are directed towards enhancing the capability of governments to deploy artificial intelligence, the outcomes of the study empirically validate and substantiate Research Hypothesis H2, which posits a positive effect of such spending on AI government readiness in the European Union countries. ECONOMIC FREEDOM is positively significantly associated with the government’s ability to implement artificial intelligence (Ciftci & Durusu-Ciftci, 2022; Le & Kim, 2020; Okulich-Kazarin et al., 2020). The more economic freedom is experienced in EU countries, the more it creates premises for governments to adopt artificial intelligence. Drawing from the empirical findings of the study, Research Hypothesis H3, which posits that there is a positive effect of economic freedom on AI government readiness in the European Union countries, has been validated. The financing by the European Commission of research & development firms for artificial intelligence projects is also found to be a significant determinant of increasing artificial intelligence in public administration (Spence, 2021). The higher the percentage of research & development companies that benefit from financing artificial intelligence projects, the more national governments benefit from scientific research results in artificial intelligence, which 284 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... can be developed and implemented at government level. Upon rigorous examination of the data, Research Hypothesis H4, which postulates a positive effect of EU-funded projects for R&D on AI government readiness in the European Union countries, has been validated. In the instrumental-variables regressions, GOVERNMENT EXPENDITURE is considered the instrumented variable, whose configurated instruments are lnGDP, GOVERNANCE and AI BACHELOR. Employing GDP as an indicator to assess the influence of AI on the economy has been beneficial for gauging governmental AI readiness. GDP, a prevalent metric of economic activity, when used to measure AI’s impact, underscores the technology’s significance for economic expansion. By monitoring AI’s contribution to GDP, governments can achieve a more profound comprehension of AI’s economic benefits and identify the sectors where AI can make the most substantial impact (Piasecki et al., 2021). The development of Government Pillar of AI Index has been a positive development for government AI readiness. The index is a tool for assessing the readiness of governments for AI adoption, and it includes a range of metrics related to AI policies, regulations, and infrastructure. It provides the assessment of vision, governance and ethics, digital capacity, and adaptability of governments in AI implementation. By using the index to benchmark their AI readiness, governments can identify areas for improvement and develop strategies to support the growth of AI in their countries. The Government Pillar of AI Index also provides a platform for international comparison, allowing governments to assess their AI readiness in relation to other countries. This can help to promote competition and encourage governments to adopt policies and strategies that support the growth of AI in their countries. As AI’s role in shaping the future of work expands, its integration into University Bachelor’s Programmes has gained importance. Governments globally acknowledge the significance of AI readiness and developing AI strategies for sustainable economic growth. The use of GDP and AI in University Bachelor’s Programmes, as well as the development of Government Pillar of AI Index, have both had a positive impact on government AI readiness. The drive to integrate AI into University Bachelor’s Programmes stems from the increasing demand for AI skills in the job market. The application of AI tools is quickly becoming an integral part of numerous industries, including finance, healthcare, and manufacturing. Universities, by incorporating AI in their programmes, equip students with vital skills for the future workplace (Vicsek, 2021). Moreover, this integration fuels innovation and economic growth. AI is a potent innovation tool, and the emergence of new AI technologies could stimulate new industries and jobs. Universities, by teaching students the skills needed to create and apply AI solutions, spur economic growth and pave the way for new opportunities for graduates. In summary, the use of GDP as an AI impact measure, the creation of the Government Pillar of the AI Index, and the integration of AI into University Bachelor’s Programmes have positively influenced government AI readiness. By tracing AI’s economic benefits, providing a gauge for government AI readiness, and preparing students with essential skills for the future workplace, these initiatives bolster global governments’ capacity to leverage the opportunities AI offers. 4.2. Cluster analysis To establish groups of EU countries and discover their mutual similarities based on the set of independent variables associated with the government artificial intelligence readiness, we Journal of Business Economics and Management, 2024, 25(2), 268–296 285 carried out the cluster analysis for the year 2022. EU countries are characterized by disparities in socio-economic development and unequal national evolution, especially as the states that joined the European Union in 2004 and 2007 (Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia) experienced centralized communist economies until the early 1990s, have since passed through difficult periods of transition to the market economy and have had to overcome structural differences and gaps in their economies compared to those of the developed countries of the European Union. In addition, traditionally, national inequalities also derive from the different size of states, discrepancies in equipment and public technological infrastructure. The determination of homogeneous clusters of states in terms of AI implementation in government is quite challenging given the obvious conditions of heterogeneity exposed. The clusters associated with the government artificial intelligence readiness index were developed through two categories of hierarchical clustering algorithm (complete linkage that produces spatially compact clusters and Ward’s linkage where groups were joined to maximize an error-sum-of-squares objective function), which were applied to establish the groups of homogeneous states and which creates a hierarchy of clusters, based on dissimilarity measure in comparing two observations. The suitable number of clusters was established based on cluster stopping rules (Duda-Hart pseudo-T-squared small values and Calinski-Harabasz pseudo-F large values, which both indicated four clusters), as well as from the study of the dendrogram and graphic representation. By applying the complete linkage and Ward’s methods for year 2022, four countries clusters based on their similarities are obtained (Table 5). The synthetic description of the characteristics of the countries in each cluster, according to the studied variables, confirms that the degree of implementation of artificial intelligence in public services is inversely proportionally with the brain drain phenomenon. Also, the positive associations of the dependent variable with spending expenses, economic freedom and research and development projects funded by the European Commission for AI companies are confirmed. Table 5. Clusters by AI IN GOVERNMENT – complete linkage and Ward’s methods Group of country AI IN GOVERNMENT dimension Number of cluster – complete linkage method Number of cluster – Ward’s method Countries with the average highest degree of government artificial intelligence Czechia, Finland, Netherlands, Luxembourg, Austria, Denmark, Estonia, Sweeden High to medium (in terms of AI IN GOVERNMENT) Medium to low (in terms of BRAIN DRAIN) High (in terms of ECONOMIC FREEDOM and AI R&D FIRMS EC FUNDED PROJECTS) High to medium and low (in terms of GOVERNMENT EXPENDITURES) Cluster 1 Cluster 1 Countries with high level of government artificial intelligence France, Germany High to medium (in terms of AI IN GOVERNMENT) Medium (in terms of BRAIN DRAIN, ECONOMIC FREEDOM and AI R&D FIRMS EC FUNDED PROJECTS) High (in terms of GOVERNMENT EXPENDITURES) Cluster 2 Cluster 3 286 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... Group of country AI IN GOVERNMENT dimension Number of cluster – complete linkage method Number of cluster – Ward’s method Countries with moderate level of government artificial intelligence Belgium, Hungary, Italy, Greece, Portugal, Slovenia, Spain Medium (in terms of AI IN GOVERNMENT) Medium to low (in terms of BRAIN DRAIN) High to medium (in terms of AI R&D FIRMS EC FUNDED PROJECTS) High to medium and low (in terms of GOVERNMENT EXPENDITURES) Medium to low (in terms of ECONOMIC FREEDOM) Cluster 3 Cluster 2 Countries with low degree of government artificial intelligence Romania, Slovakia, Croatia, Latvia, Poland, Bulgaria, Lithuania High to medium (in terms of AI IN GOVERNMENT) High (in terms of BRAIN DRAIN) Medium to low (in terms of GOVERNMENT EXPENDITURES, ECONOMIC FREEDOM) Low (in terms of AI R&D FIRMS EC FUNDED PROJECTS) Cluster 4 Cluster 4 The geographical representation of the countries based on the main proxy BRAIN DRAIN compared with the cluster’s representation in complete linkage methods are illustrated in Figure 1. The lowest values of core explanatory variable (BRAIN DRAIN) are in Romania, Bulgaria, Croatia, Slovakia, Poland, Latvia, Lithuania and Estonia. This group of states overlaps (with the exception of Estonia) with the last cluster obtained, cluster 4, determined on the basis of cluster analysis and containing states with low degree of government artificial intelligence. The similar type of correspondence between the two maps (Figure 1) is found for the first three clusters, in the sense that the countries in categories 2, 3 and 4 after the decreasing volume of the brain drain are in distinct clusters and for the entire set of data analysis by the cluster method, in which all variables were considered in the AI government estimation (Figure 1b and Table 6). For example, the countries with the lowest brain drain values (Finland, Figure 1. Geographical representation of the EU (24) countries based on: a – brain drain; b – Clusters in complete linkage method a) b) End of Table 5 Journal of Business Economics and Management, 2024, 25(2), 268–296 287 Netherlands, Luxembourg, Denmark, and Sweden) are included in cluster 1 with the average highest degree of government artificial intelligence. States such as Czechia and Austria, placed in cluster 1 and having medium brain drain values are in the first part of the states studied from the perspective of adopting AI in government. Estonia’s case is a special one, in the sense that although it presents high values of brain drain migration flows, it has developed and consolidated AI in government, through targeted and successful national measures. Germany and France (cluster 2 in complete linkage method) present medium values in terms of brain drain and high level of government artificial intelligence, while the rest of countries (cluster 3 – Belgium, Hungary, Italy, Greece, Portugal, Slovenia, Spain) shows average to low brain drain values, as well as average values of AI in government. The analyzed geographical distribution reveals situations in which immediately neighboring states have the same type of behavior from the perspective of the clusters of belonging and the variation of the variables studied by the cluster analysis, which adds preliminary signals of influences between states, as we will study in the following through spatial analysis. To determine if the selected independent variables have the potential to explain the evolution of the degree of implementation of the government artificial intelligence, we estimate a cluster multivariate regression model (Table 6). The results display high rates of explanation of the chosen model for each independent variable (R-squared), as well as statistically significant outcomes of the F statistic, which indicates that the model has predictive capability. Table 6. Clusters analysis for AI IN GOVERNMENT (source: author's calculations using STATA) AI IN GOVERNMENT F R-squared BRAIN DRAIN 13.596*** 0.671 GOVERNMENT EXPENDITURE 29.465*** 0.815 ECONOMIC FREEDOM 10.468*** 0.610 AI R&D FIRMS EC FUNDED PROJECTS 21.356*** 0.762 Note: * Significant at 10%; ** Significant at 5%; *** Significant at 1%. 4.3. Spatial analysis After identifying the relationship between the variables studied by the instrumental-variables regression method and cluster countries analysis, another aim of the study is to find the spatial pattern of the data, more precisely whether AI IN GOVERNMENT and its determinants display agglomeration of high values with high values between countries, low values with low values or high values is grouped next to low values. To perform the spatial analysis, we initially design a spatial weight matrix in the GeoDa software (Anselin et al., 2006), which defines the neighborhoods between states, in a binary format, in which if there is a common border between countries the value is 1 and otherwise 0. As the weight matrix cannot reflect the specific relationships between countries (human, economic or governance in our case), we develop the nested spatial weight matrix obtained by multiplying the adjacent matrix with the weight matrix for human, economic or governance distance from our data. We use the nested standardized matrix, which can reflect the spatial characteristics of neighboring countries, as well as the degree of interdependence between countries (Elhorst, 2017; Wang et al., 2023). Based on nested matrix, we estimate the global Moran index I and obtain positive values of it and quite close to 1 for the explained and core explanatory variables (AI IN GOVERNMENT 288 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... and BRAIN DRAIN) which indicates that these variables have a substantial positive spatial autocorrelation, and the data is grouped on the principle high-high or low-low (Table 7). Values are also positive, but closer to 0 are obtained for control variables and show a lower degree of spatial grouping of observations. Overall, data presents significant spatial correlation and is suitable for their use in spatial models. Table 7. Global Moran’s I spatial correlation for the selected variable, year 2022 Variable(s) Moran’s I p-value z-value AI IN GOVERNMENT 0.479 0.002 2.818 BRAIN DRAIN 0.699 0.000 3.930 GOVERNMENT EXPENDITURE 0.193 0.071 1.469 ECONOMIC FREEDOM 0.258 0.054 1.612 AI R&D FIRMS EC FUNDED PROJECTS 0.267 0.048 1.666 Note: * Significant at 10%; ** Significant at 5%; *** Significant at 1%. Standard errors are in round brackets. Following the confirmation of the existence of spatial correlations, the spatial error regression model is examined. The robustness check of the model shows that the lambda parameter is significant (as are the Wald and Lagrange multiplier tests), and the values of variance ratio and squared correlation ensure a considerable predict of dependent variable through explanatory indicators. From the empirical results (Table 8) we can find that both core explanatory variable and control variables present statistically significant values of spatial correlation coefficients and explain government AI from spatial perspective between countries. The increasing BRAIN Table 8. Spatial Error Model regression results (source: authors’ calculations using STATA) AI IN GOVERNMENT SEM BRAIN DRAIN –1.826*** (0.633) GOVERNMENT EXPENDITURE 0.008** (0.003) ECONOMIC FREEDOM 0.703*** (0.108) AI R&D FIRMS EC FUNDED PROJECTS 22.039* (11.850) Constant 5.595 (14.669) Variance ratio 0.843 Squared correlation 0.825 Lambda –0.466*** (0.127) Wald test 13.378*** Lagrange multiplier LM test 4.594** Note: * Significant at 10%; ** Significant at 5%; *** Significant at 1%. Robust standard errors are in round brackets. Journal of Business Economics and Management, 2024, 25(2), 268–296 289 DRAIN in neighboring countries has a spatial spillover effects or negative externality on the home country’s AI IN GOVERNMENT. Also, the improvement of macroeconomic conditions (GOVERNMENT EXPENDITURE and ECONOMIC FREEDOM), as well as those for financing AI R&D (AI R&D FIRMS EC FUNDED PROJECTS) in neighboring countries are responsible for increasing the capacity of governments to manage AI adoption. Spatial analysis of the influence of brain drain on AI in government highlights regional and global inequalities in the development and deployment of AI technologies in the public sector. Addressing these inequalities requires concerted efforts by governments, academia and the private sector to develop a strong talent base in AI, encourage international collaboration and support the development of local AI solutions tailored to the specific needs of different governments. Spatial analysis of the influence of the brain drain on AI in government highlights regional and global disparities in the development and implementation of artificial intelligence in the public sector. Brain drain refers to the migration of highly qualified specialists, including AI experts, from countries of origin to destinations with better opportunities, such as developed countries (Kapur & McHale, 2005). This talent migration can have a negative impact on governments’ ability to develop and deploy AI solutions. Regional imbalances in AI development can critically affect government efficiency in public service delivery and policy formation (Wirtz et al., 2020). Governments with AI talent access and the resources to realize AI solutions are better equipped to tackle complex social and economic problems, meeting citizen needs (Mikhaylov et al., 2018). Conversely, governments failing to attract and retain AI specialists could see their public service and policy management efficiency decline (Zajko, 2022). The brain drain can exacerbate inequalities between countries in access to innovations and advanced technologies in AI. Countries that manage to attract and retain talent in AI can benefit from investment in R&D and accelerated economic growth (Singh & Krishna, 2018). The concentration of talent in AI in certain regions or countries can amplify differences in AI development and adoption in governments around the world. Developed countries, especially Western European countries, have managed to attract many AI specialists, strengthening their AI positions. In contrast, developing countries and local governments may face difficulties in attracting and retaining AI talent (Banerjee & Duflo, 2019). At the same time, countries affected by the brain drain may lag behind in the technological race, which may limit their ability to address social and economic problems and respond to citizens’ needs (Sabry, 2021). To address these spatial inequalities and encourage a more equitable development of AI in the public sector, governments need to consider several strategies. First, investment in education and research in AI should be encouraged and supported at local and regional level (Basri & Box, 2010). This can help develop a solid talent base in AI and create employment opportunities in the public sector (Yamashita et al., 2021). Second, international AI collaboration can be crucial to combat the brain drain and ensure that former communist EU countries have access to cutting-edge knowledge and technologies (Berger, 2022). Partnerships between countries and institutions can facilitate the transfer of technology and knowledge and contribute to the development of AI solutions tailored to the specific needs of different governments (Qin et al., 2023). Third, governments can encourage the development of local AI solutions and support small and medium-sized enterprises (SMEs) specializing in AI. Creating an innovative AI ecosystem can attract local and international talent and help build governments’ capacities to implement AI in the public sector (Djeffal et al., 2022). 290 I. C. Iuga, A. Socol. Government Artificial Intelligence readiness and brain drain: influencing factors and spatial effects... The findings lead to various inferences and recommendations. For public administrations, AI extends beyond being merely instrumental; it’s a transformative catalyst with potential to reshape governance, policy formation, and public service provision. AI can help governments generate accurate predictions and simulate complex systems to experiment with different policy options (Margetts & Dorobantu, 2019). Hence, it’s crucial for public servants to keep pace with AI advancements and comprehend its applicability for societal welfare (Reis et al., 2019). Moreover, AI’s role in government underscores the necessity for robust collaborations amongst public institutions, private enterprises, academic bodies, and civil society organizations. Such partnerships encourage knowledge exchange, resource sharing, and expertise pooling, which are vital for crafting AI systems tailored to governmental needs and goals. It is important to note that brain drain impedes government AI development, as the emigration of AI professionals slows AI progress in public services, reducing expertise and innovation. This results in economic losses for governments, especially in resource-limited nations, and hampers international AI collaboration due to less cooperation from relocated experts. Such nations risk falling behind in the global AI race, while those retaining talent could advance technologically and economically. Government spending is key to AI readiness, and stable economic policies favor AI adoption. The European Commission’s AI R&D investment underscores the need for ongoing financial support in this field. Finally, addressing brain drain is vital for a nation’s AI future, economic growth, and innovation. Governments must tackle these challenges with proactive strategies, fostering an environment conducive to AI progress for the benefit of all citizens. 5. Conclusions Given the intersection between Artificial Intelligence (AI) and brain drain, this study garners broad public interest by providing valuable insights, enhancing understanding of AI’s impact on society, economy, and environment and highlighting the need for governments to regulate and implement AI across various sectors. This study offers nuanced understandings of the factors influencing AI government readiness within the European Union. Specifically, it delves into the potential negative repercussions of Brain Drain (RQ1) and the potential positive implications of EU-funded projects for R&D, economic freedom, and government spending (RQ2) on AI preparedness. Moreover, the research provides crucial insights into the interplay between brain drain and a country’s influence on its neighboring states’ AI adoption readiness (RQ3). These findings elucidate the significance of understanding both internal and external dynamics to enhance AI readiness and integration at the governmental level in the European region. Our findings unequivocally indicate a negative correlation between brain drain and AI readiness in EU governments in 2022. This suggests that as brain drain intensifies, governments’ preparedness to integrate AI diminishes. This correlation, coupled with the insights garnered from the spillover effects and the potential policy interventions, offers a comprehensive roadmap for EU nations to strategically address the challenges posed by brain drain in the realm of AI readiness. Several implications arise from our findings: Detrimental Effects of Brain Drain: Brain drain significantly undermines government AI readiness. Skilled AI professionals migrating abroad impede the progress and application of advanced AI in public services, confirming Hypothesis H1 and supporting existing research findings. Journal of Business Economics and Management, 2024, 25(2), 268–296 291 Innovation Lag: The migration of AI talent leads not only to a loss of expertise but also diminishes the potential for technological innovation within their home countries. Economic Implications: When AI experts leave, their home countries lose the investments made in their education and training. This exacerbates economic difficulties, especially in resource-constrained nations. Challenges in International Collaboration: The brain drain phenomenon can adversely affect international AI collaborations. Experts relocating abroad might become less willing to engage with their home countries, resulting in lost opportunities in international AI projects. Global AI Race: Countries severely affected by brain drain risk falling behind in the international competition for AI advancement. In contrast, those retaining or attracting AI talent could see significant technological and economic growth. Prominence of Government Expenditure: There is a notable correlation between government investment and AI readiness. Adequate government funding can substantially improve the public sector’s AI capabilities. The Role of Economic Freedom: An enabling economic environment is crucial for the adoption of AI in government. Countries with greater economic freedom typically have an easier path to AI integration, emphasizing the importance of stable and supportive economic policies. European Commission’s AI R&D Investment: The European Commission’s funding in AI research and development has been pivotal. This investment is essential for advancing AI readiness, highlighting the importance of ongoing financial support in this field. Addressing brain drain is crucial for a nation’s future in the AI arena, ensuring economic growth and innovation. Governments need strategies to retain talent and create AI-friendly environments, benefiting citizens. However, our study faces limitations, including data availability constraints. Despite using reliable sources, data gaps and discrepancies may exist due to AI’s dynamic nature and fluctuating socio-political factors influencing brain drain. Additionally, incorporating diverse qualitative insights, like AI professionals’ or policymakers’ experiences, could have enriched our methodology. Future research could delve deeper into the following areas: Global Perspective (while our study focused on the EU, understanding this dynamic in other regions or on a global scale could provide more comprehensive insights); sectoral Analysis (AI’s integration varies across sectors. Future studies could examine how brain drain affects AI readiness in specific government sectors, such as health or defense). Author contributions The authors (ICI and AS) contributed equally to the elaboration of this research. Disclosure statement Authors have no competing financial, professional, or personal interests from other parties. References Agrawal, A., Gans, J., & Goldfarb, A. (2019). The economics of artificial intelligence: An agenda. University of Chicago Press. https://doi.org/10.7208/chicago/9780226613475.001.0001