The finding comes from an International Monetary Fund working paper analysing more than 700 million online job advertisements across 110 economies between November 2021 and April 2026.
Researchers compared hiring trends in occupations with high exposure to AI against those less affected following the launch of ChatGPT in November 2022.
Job advertisements in highly exposed occupations declined by nearly 10% relative to less exposed roles. The reduction emerged in late 2023 and grew during 2024 before stabilising.
The pooled estimate across the period showed hiring in highly exposed occupations falling by about 6%.
However, the decline did not deepen as AI systems became capable of completing more complex and time-consuming tasks.
Researchers said the stabilisation could reflect companies redesigning jobs and production processes to incorporate AI rather than continuing to reduce recruitment.
The working paper identified 51 emerging AI skills, including familiarity with ChatGPT and Copilot, prompt engineering, AI agents, vector databases and retrieval-augmented generation.
Demand for those skills has grown rapidly since 2023 and is concentrated in occupations most exposed to AI.
Companies advertising positions requiring emerging AI skills recorded faster growth in overall job postings than businesses not seeking those capabilities.
That pattern remained evident even in occupations considered highly exposed to automation.
“AI appears to be reshaping labour demand by both compressing hiring in more exposed occupations and generating new opportunities for workers with AI skills,” the researchers said.
The negative effect was strongest in advanced economies, smaller in emerging markets and statistically insignificant in low-income economies and Africa.
The study found no clear difference between junior and senior hiring and no evidence that AI exposure had affected advertised wages.
Researchers cautioned that online job advertisements primarily reflected formal-sector demand and could understate changes occurring through informal employment, particularly in developing economies.
The findings suggest that workers’ prospects will increasingly depend on acquiring skills that complement AI as businesses adapt jobs and workflows.
–IMF/ChannelAfrica–
