AI Exposure and Preparedness in Slovakia’s Labor Market1
This analysis uses measures of exposure and complementarity to Artificial Intelligence (AI) on employment data from Slovakia to examine the potential implications of AI on the labor market. Around half the workers in Slovakia have high exposure to AI, with significant variations across sectors. While 27 percent of jobs also exhibit high complementarity, 24 percent are at risk of displacement especially in certain service sectors such as finance and information and communications. Women in particular have higher exposure to AI, and face both greater opportunities and greater risks of job displacement from AI. Slovakia has relatively low AI preparedness and digital skills compared to its peers. Policies to support a digital ecosystem, enhance human capital and manage labor market transitions, as well as a supportive regulatory framework would allow Slovakia to maximize the opportunities and mitigate the negative impacts from AI.
A. Introduction
1. Artificial Intelligence (AI) holds the potential to significantly impact economies globally, with profound shifts particularly in the labor market. AI has the potential to enhance productivity and boost growth – it could generate a positive productivity shock, which broadens the productive capabilities of economies, and facilitate shifts between labor and capital. The advent of generative AI (GenAI) in particular, with its cognitive capabilities, has broadened AI’s potential applications. In some economic sectors and jobs, AI could augment worker productivity and boost labor demand; conversely, other sectors could see large job displacements if AI reduces the need for human input. However, given the rapid evolution of the AI landscape, its vast and flexible applications in numerous domains, and society still grappling with the acceptability of its use, the impact of AI on economies and societies remain highly unpredictable.
2. The impact of AI is likely to vary significantly across countries, depending on the level of development and economic structure. Advanced economies (AE) are likely to reap greater benefits and experience more of the negative effects of AI on the labor force compared to emerging markets and developing economies (EMDEs), primarily due to their workforce being more oriented towards cognitive-intensive jobs. This could exacerbate economic disparities across countries. While AI-induced job displacement poses risks for higher-wage earners, potential AI complementarity is also correlated with income, which could affect income and wealth inequalities within countries.
3. This analysis studies the potential impact of AI on the labor market in Slovakia. It aims to link the findings from the academic literature onto the specific circumstances of the Slovak labor market, looking at the impact on various population groups and economic sectors.
B. Conceptual Framework
4. The analysis uses a conceptual framework that measures various jobs’ exposure to, and complementarity with, AI. It uses the occupational classification proposed by Cazzaniga et al (2024), which is in turn based on the work of Felten et al (2021) and Pizzinelli et al (2023). Exposure is defined as the degree of overlap between AI applications and required human abilities in each occupation (Felten et al, 2021), also known as the standard measure of AI occupational exposure (AIOE). Pizzinelli et al (2023) augment this measure with an index of potential AI complementarity, by considering information on the social, ethical, and physical context of occupations, as well as required skill levels. This enables the framework to account for the potential of AI as a labor complement or substitute, where complementarity reflects the degree of shielding from AI-driven job displacement. Paired with AI exposure, this framework can thus give an indication of the amount of jobs at risk of being made redundant by AI.
5. Using this framework, occupations can be categorized according to their exposure to, and complementarity with, AI. We categorize jobs into four groups in the exposure/complementarity quadrant: high exposure/high complementarity (HE/HC), high exposure/low complementarity (HE/LC), low exposure/high complementarity (LE/HC) and low exposure/low complementarity (LE/LC). These categories are determined based on whether the exposure and complementarity to AI of occupations fall above or below the respective median values (Figure 1). HE/HC occupations are those that stand to benefit the most from AI, including because there is significant potential for AI support and limited scope for unsupervised use (for example due to ethical or societal concerns). These are primarily cognitive jobs with a high degree of responsibility and personal interactions (e.g., surgeons, judges). Meanwhile, HE/LC occupations are well-positioned for AI integration but with a greater likelihood of AI replacing human tasks (e.g., telemarketers) – and hence most vulnerable to displacement from widespread AI adoption.
AI Exposure and Complementarity
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
C. Labor Market Implications for Slovakia
6. The conceptual framework is applied to the Slovak labor market to assess AI exposure and complementarity. Using the data by occupational classifications mentioned above, AI exposure and complementarity was estimated for Slovakia by applying the employment data (ISCO 3-level digit) provided by the Statistical Office of the Slovak Republic (SOSR). The employment data are 2023 yearly averages obtained from the quarterly labor force survey and include self-employed individuals. This allows for a more granular look at the potential impact of AI on the distribution of occupations in Slovakia.
7. Several caveats apply. Importantly, the level of complementarity and exposure estimated is based on US data, and applying the classification framework assumes that tasks performed within similar occupations are homogeneous around the world, thus ignoring likely cross-country variations. For example, jobs with higher complementarity require higher cognitive skills, but the PIAAC (Program for the International Assessment of Adult Competencies) data suggests that Slovak workers tend to use such skills less often than US workers, suggesting that the results in our analysis might overestimate the share of high exposure and high complementarity jobs in Slovakia (Figure 2). The large gap in the use of cognitive skill by clerks and service workers between the US and Slovakia also suggests such workers could face higher risks of displacement from AI in Slovakia compared to their US counterparts. The main analysis in this paper relies on the assumption of homogeneity of tasks to simplify international comparisons. However, we also show adjusted results for Slovakia based on the PIAAC data (see Box 1). Another important caveat is that that the analysis is static, assuming that sector sizes and tasks in each occupation remain unchanged even after the introduction of AI.
Use of Skills at Work
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
Note: influencing skills ranges from 1 to 5, reading ranges from -3 to 7, and task discretion ranges from -1 to 5Sources: OECD and IMF staff calculation.Adjusting for Skill Differences Between Slovakia and the US
This box adjusts the complementarity index used in this paper using PIACC data to better reflect the differences in skill usage between the US and Slovakia. The analysis in this paper assumes that the tasks performed in the US are homogeneous around the world by using the same occupation classification. However, as shown in Figure 2, the skills used at work can differ globally, even within the same occupation category. To address this issue as much as possible, the data for complementarity is adjusted based on the PIAAC data (2nd Cycle) for the US and Slovakia.1
The adjusted data implies that Slovakia could face higher risks of AI displacement compared with the US. Comparing the data before and after the adjustment, the share of HE/HC jobs in Slovakia — occupations which are likely to benefit the most from the introduction of AI— decreases from 27 to 17 percent. On the other hand, the share of HE/LC jobs increases to 34 percent from 24 percent in the unadjusted data. This suggests that job disruptions from AI in Slovakia could be larger than suggested by the unadjusted index.
1 The complementarity estimated by Pizzinelli et al. (2023) using O*NET consists of six components. Based on the percentage difference in the average skill usage between US and Slovak workers, each of the six components is adjusted at the level 2 of ISCO. However, if the sample size of a certain category is less than 30, the data for ISCO level 1 is used instead. The PIAAC variables used for the adjustment are as follows: 1) Communication: F2_Q01b, F2_Q02a , F2_Q05a , H2_Q03c, 2) Responsibility: H2_Q03b, H2_Q03d, H2_Q14a, 3) Physical Conditions: H2_Q07a, H2_Q07b, 4) Criticality: H2_Q05a, H2_Q05b, H2_Q06b, 5) Routine: TASKDISCC2_T1, 6) Skills: D2_Q12a , D2_Q12d. See OECD (2024) for detailed data and codebooks.8. The share of employment with high AI exposure in Slovakia is slightly lower than in other advanced economies (AEs) (Figure 3). Around half the labor force in Slovakia is highly exposed to AI, with 27 percent of jobs characterized by high complementarity (HE/HC) and 24 percent at risk of AI-related displacement (HE/LC). In general, AEs have a larger share of highexposure occupations (57 percent) than emerging market economies (EMs, 33 percent) and low-income countries (LICs, 24 percent). Slovakia’s share of HE/HC occupations is similar to the average AE (27 percent of employment in SVK and 26 percent in AE), while the share of HE/LC high exposure/low complementarity occupations is slight lower (24 percent in SVK and 32 percent in AE). Compared to other EU countries, the share of HE/HC occupations is lower, and the share of high HE/LC occupations slightly higher.
Employment Shares by AI Exposure and Complementarity: Country Groups and Selected European Countries
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
9. Service sector jobs in Slovakia stand to gain the most from widespread AI adoption, but workers in these jobs are also at the highest risk of displacement (Figure 4).
HE/HC: A large share of the jobs in the education, wholesale and retail, real estate, and professional sectors exhibit high exposure and high complementarity to AI. These jobs are well-positioned to take advantage of productivity gains from AI and emerging AI growth opportunities.
HE/LC: A significant proportion of finance, information and communication, and public sector jobs exhibit high exposure and low complementarity. In Slovakia, about two-thirds of jobs in finance (e.g. financial professionals), and information and communication (e.g., software and application developers and analysts) exhibit high AI exposure and low complementarity, making them more susceptible to risks from AI-related labor market shifts. Public sector occupations also fall into this category, with around half of the jobs in this sector (e.g., general office clerks) at risk.
10. Jobs in the manufacturing sector, which accounts for the highest share of employment in Slovakia, are relatively less exposed to AI. Around 70 percent of manufacturing employment has low AI exposure. Only 20 percent of manufacturing jobs (e.g., physical and engineering science technicians, clerical support in manufacturing) have high exposure and low complementarity to AI and thus at greater risk of job displacement. Applying the analysis to the automotive sector specifically, the results are very similar, although the small sample of data available makes it less reliable. However, these jobs might face the threat of replacement by robots or automation instead of AI, due to the nature of their manual tasks, although this is beyond the scope of this paper.
AI Exposure and Complementarity by Sector and Gender
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
11. Female workers are more exposed to AI than male workers in Slovakia. 64 percent of female workers in Slovakia have high exposure to AI, while for men, the share is 39 percent.2 In Slovakia, more women are employed in high-exposure occupations such as clerical, services and sales, and professional occupations, while there is a higher share of men in low-exposure jobs such as crafts and machinery operation occupations (Figure 4). This is similar to the finding in Cazzaniga et al (2024) where in most countries, more women tend to be employed in high-exposure jobs relative to men. By complementarity, women with HE/HC jobs make up 35 percent of total female employment, relative to 20 percent of men in Slovakia. Conversely, almost 30 percent of females are in the HE/LC category compared to 20 percent of men, which reflects the higher share of women in certain jobs (especially clerical jobs). This can be interpreted to mean that women face both higher risks and higher opportunities, consistent with the finding in Cazzaniga et al (2024) for most countries.
12. The degree of exposure and complementarity to AI is broadly similar across age groups, and higher for more educated workers (Figure 5). The distribution of jobs according to AI exposure and complementarity is relatively equal across age groups in Slovakia. In terms of education level, more educated workers have jobs with high AI exposure, and many of those jobs are in occupations with high AI complementarity. This is similar to the finding for other countries (Cazzaniga et al, 2024).
AI Exposure and Complementarity by Age and Education
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
D. AI Preparedness and Digital Skills
13. The level of AI preparedness in Slovakia is below the average of EU countries (Figure 6). The AI preparedness index (AIPI) developed by Cazzaniga et al (2024) looks at four key dimensions relevant for smooth AI adoption: (i) digital infrastructure; (ii) innovation and economic integration; (iii) human capital and labor market policies; and (iv) regulation and ethics.3 While there is high uncertainty around the institutional requirements for an economy-wide integration of AI, the AIPI can point toward areas for improvement. The AIPI for Slovakia is 0.6, lower than the average EU country (0.7) and the average AE (0.7). Although performing better than EMs, Slovakia scores at the low end of the AE grouping (Figure 6). In particular, Slovakia lags in regulation and ethics, and innovation and economic integration relative to the average EU and AE country.4
AI Preparedness
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
14. Digital skills and the use of information, communication, and technology (ICT) in Slovakia is also generally lower than the EU median (Figure 7). Basic or above basic digital skills in Slovakia are below the EU median across all types of digital skills, as well as across all age groups and gender. In terms of occupations, digital skills of manual jobs are at the EU median, while other jobs are below. In addition, Slovak firms are below the EU median in terms of advanced ICT usage. While small firms are on par with the median in terms of AI usage, medium and large Slovak firms lag behind. By sector, the information and communications firms are below the EU median on AI usage.
Digital Skills and ICT/AI usage
Citation: IMF Staff Country Reports 2025, 073; 10.5089/9798229005852.002.A003
E. Conclusions and Policy Considerations
15. While the impact of AI on economies and the implications on society are challenging to predict, it has the potential to have large impacts on the labor market. The high uncertainty about its socioeconomic implications can be linked to its rapid development and fast-evolving landscape, and also its nature as a general-purpose technology, akin to electricity and the internet. AI could cause potential disruptions in the labor market, bringing gains via a productivity boost to some workers, and also risks by displacing jobs in certain industries.
16. A substantial share of the labor force in Slovakia will be impacted by AI, particularly female workers. Around half of all Slovak workers are highly exposed to AI, including almost a quarter of whom are at risk of job displacement from widespread AI adoption. The rest of the highly exposed workers – 27 percent in Slovakia – have jobs with high complementarity to AI and are thus likely to benefit from widespread AI adoption. In terms of sectors, some services could face higher risks of AI displacement, e.g., finance, information and communication, and professional and scientific sectors. Female workers in Slovakia in particular have a higher degree of exposure to AI. They stand to reap the potential benefits of AI adoption but are also more at risk from AI-related job displacement.
17. Policymakers can take action to ensure readiness to AI adoption, to harness the benefits and manage the risks of labor market displacement.
Support the development of a digital ecosystem that is conducive to AI adoption. Greater R&D spending in relevant areas can foster technological advancements and enable the smooth integration of AI through the economy. The use of AI and advanced ICT is relatively low amongst Slovak firms relative to EU peers, and improving the digital ecosystem could help in increasing the adoption of AI and digital skills.
Human capital and labor market policies to ensure a skilled labor force and to manage AI-induced job transitions:
Enhance human capital through education and training. Policies could be taken to ensure that the education system is able to prepare the workforce for AI and digital literacy, and is agile to help workers adapt to new technologies.
Policies must promote an equitable and ethical integration of AI and ensure social cohesion. This could include supporting business in integrating AI responsibly.
Support workers through transitions – reskilling or upskilling. Policies should facilitate the reallocation of labor while providing support for workers that are affected by AI-induced transitions.
References
Cazzaniga, M., Jaumotte, M.F., Li, L., Melina, M.G., Panton, A.J., Pizzinelli, C., Rockall, E.J. and Tavares, M.M.M., 2024. “Gen-AI: Artificial Intelligence and the Future of Work.” IMF Staff Discussion Note SDN2024/001, International Monetary Fund, Washington, DC.
Felten, E., M. Raj, and R. Seamans. 2021. "Occupational, Industry, and Geographic Exposure to Artificial Intelligence: A Novel Dataset and Its Potential Uses." Strategic Management Journal 42 (12): 2195–217.
OECD (2024) "Programme for the International Assessment of Adult Competencies, 2nd Cycle Database", https://www.oecd.org/en/data/datasets/piaac-2nd-cycle-database.html (accessed on 24 January 2025)
Pizzinelli, C., A. Panton, M. M. Tavares, M. Cazzaniga, and L. Li. 2023. "Labor Market Exposure to AI: Cross-Country Differences and Distributional Implications." IMF Working Paper 2023/216, International Monetary Fund, Washington, DC.
Prepared by Shinya Kotera and Yen Mooi (EUR).
Females comprise 47 percent of total employment in Slovakia and males are 53 percent.
The index is composed of a set of indicators expected to be important for smooth AI adoption, including sustained human capital investment, inclusive expertise in STEM (science, technology, engineering and mathematics), labor and capital mobility, and adaptability of legal frameworks to digital business models. The indicators are normalized to a 0-1 scale and averaged, with the AIPI a simple average of the four dimensions.
These two indicators (innovation and economic integration, and regulation and ethics) can be considered “second-generation” elements likely to maximize the economic impact of AI. Digital infrastructure and human capital and labor market policies can be considered “foundational” elements, more relevant for AI adoption. (Cazzaniga et al, 2024)
