Mapping AI’s Labor Impact: A Task Exposure Framework for Occupational Analysis

The rise of Artificial Intelligence (AI) is transforming labor<br/>markets globally, prompting a re-evaluation of job tasks and required skills.<br/>This paper investigates AI’s potential impact on occupations, offering a<br/>technology-aware perspective on occupational exposure. Moving beyond<br/>traditional static models, our analysis utilizes Large Language Models (LLMs)<br/>to assess AI’s task-level suitability, incorporating real-world implementa-<br/>tion factors and detailed task characteristics. We introduce an extended<br/>Task Exposure to AI index (TEAI) and apply it to analyze occupational<br/>exposure patterns across sectors, skill levels, and countries. Our findings<br/>reveal highly heterogeneous impacts, with clerical support occupations ex-<br/>hibiting the highest potential for AI task transformation (up to 61% of<br/>tasks highly suitable), while service and sales roles show greater resilience<br/>(approximately 20% of tasks highly suitable). Cross-national comparisons<br/>indicate that approximately 42 million workers in Europe and 34 million in<br/>the U.S. are in occupations with high AI exposure, suggesting significant<br/>workforce transitions. Crucially, we identify substantial implementation<br/>gaps, highlighting areas where AI adoption lags behind technical feasibil-<br/>ity, and vice versa. These insights provide a robust, data-driven foundation<br/>for informing policy interventions aimed at facilitating a smoother labor<br/>market transition in the age of AI.