Modeling an AI jobs transition

OpenAI’s AI Jobs Transition Framework provides a detailed map of how artificial intelligence may affect the U.S. labor market by analyzing 921 occupations that encompass roughly 148 million jobs. Rather than predicting outright job losses, the framework classifies roles into four groups: about 18% face relatively high automation risk, 24% are likely to reorganize, 12% could grow as AI lowers costs and expands demand, and 46% show minimal immediate change. High‑risk occupations include data‑entry clerks, telemarketers, proofreaders, and certain bookkeeping or administrative positions, where AI can handle standardized tasks and demand may not offset productivity gains. Occupations expected to reorganize—such as lawyers, accountants, software developers, financial analysts, and teachers—retain human oversight for judgment, responsibility, and relationship‑building, while AI augments productivity. Growth‑potential roles like tutors, mental‑health counselors, personal financial advisers, and some health‑care professionals may see expanded employment as AI makes services more affordable and accessible. Jobs with less immediate change, such as electricians, plumbers, roofers, construction laborers, and many food‑service workers, remain largely physical and location‑bound, limiting near‑term AI exposure. The framework notes that actual AI usage, exemplified by ChatGPT, is already about three times higher in high‑risk occupations than in the workforce overall, yet recent unemployment trends do not align neatly with technical exposure, reflecting broader labor‑market dynamics. OpenAI concludes that different occupations will require distinct policy responses—early‑warning systems and transition assistance for high‑risk jobs, updated training and professional standards for reorganizing roles, and adoption‑friendly policies for growing occupations—while emphasizing the need for real‑time data to capture evolving task and technology use.