Keep Apart Research Going: Donate Today
Apr 28, 2025
Evaluating the risk of job displacement by transformative AI automation in developing countries: A case study on Brazil
Blessing Ajimoti, Vitor Tomaz, Hoda Maged , Mahi Shah, Abubakar Abdulfatah
Summary
In this paper, we introduce an empirical and reproducible approach to monitoring job displacement by TAI. We first classify occupations based on current prompting behavior from a novel dataset from Anthropic, linking 4 million Claude Sonnet 3.7 prompts to tasks from the O*NET occupation taxonomy. We then develop a seasonally-adjusted autoregressive model based on employment flow data from Brazil (CAGED) between 2021 and 2024 to analyze the effects of diverging prompting behavior on employment trends per occupation. We conclude that there is no statistically significant difference in net-job dynamics between the occupations whose tasks feature the highest frequency in prompts and the ones with the lowest frequency, indicating that current AI technology has not initiated job displacement in Brazil.
Cite this work:
@misc {
title={
Evaluating the risk of job displacement by transformative AI automation in developing countries: A case study on Brazil
},
author={
Blessing Ajimoti, Vitor Tomaz, Hoda Maged , Mahi Shah, Abubakar Abdulfatah
},
date={
4/28/25
},
organization={Apart Research},
note={Research submission to the research sprint hosted by Apart.},
howpublished={https://apartresearch.com}
}