Abstract
The rapid spread of generative artificial intelligence (AI) has raised the question of its impact on employment not only in developed but also in transition economies. The purpose of the article is to assess the social aspects of employment transformation in Kazakhstan in the context of the spread of artificial intelligence based on a comparison of employment dynamics and the level of its formalization in sectors with different exposure to AI in 2010-2024. The methodology includes an industry classification of AI exposure based on the international AI Industry Exposure index, a comparative analysis of employment dynamics, calculation of average annual growth rates, correlation analysis and intergroup statistical tests. The initial information base was made up of official data from the Bureau of National Statistics of the Republic of Kazakhstan on 17 types of economic activity for 2010- 2024. The results showed that in 2010-2024, the average employment growth was 58.5% in sectors with high exposure to AI versus 31.1% and 34.7% in sectors with medium and low exposure, respectively. After 2022, the average annual employment growth rate in highexposure sectors decreased from 3.58% to 1.98%, while in lowexposure sectors it increased from 1.33% to 2.28%. The relationship between the level of exposure to AI and the change in growth rates after 2022 turned out to be negative and statistically significant (r=- 0.53; p=0.028). The results obtained do not allow us to establish the causal effect of AI on employment, however, they record differences in industry dynamics after the beginning of the spread of generative AI and indicate the need for further analysis at the level of professions and work tasks.