OpenAI Maps Europe's AI Jobs Risk: 14% Face Automation Pressure

OpenAI's economic research team estimates that about 14% of EU employment sits in occupations with relatively high near-term automation potential, according to its new report, The AI Jobs Transition Framework for the EU, released by OpenAI Economic Research.

The report extends the AI Jobs Transition Framework, first developed for the United States in April 2026, to the European labor market. It maps AI's potential occupational effects using the official European Skills, Competences, Qualifications and Occupations (ESCO) taxonomy combined with Eurostat employment data. On the whole, the EU shows a smaller share of employment in high-automation-potential occupations than the United States.

The stakes are significant. Europe's labor market is shaped by licensing systems, local institutions, and the practical realities of delivering care, education, justice, and public services — systems that determine if and how AI changes work. The report frames the central questions bluntly: "What will AI's labor market impact be? Where will the impacts be felt most and when? And how can we ensure that the AI transition works for everyone?"

Four transition archetypes

The framework sorts occupations into four categories. It is explicit that these are not employment forecasts. "They are a planning map for where different kinds of adjustment pressure and opportunity may emerge," the report states.

Applied across EU member states, the breakdown is:

• About 12% of employment is in occupations that may grow with AI, as lower costs expand access to services or make more projects viable.
• About 14% is in occupations with relatively higher near-term automation potential.
• About 27% is in occupations likely to reorganize, where AI may change workflows and skill needs even as people remain central to delivery.
• The remaining 47% is in occupations with less immediate change.

The reorganization category is the largest of the three affected groups. It covers jobs where AI alters how work gets done without eliminating the human role — a distinction the report treats as central to realistic workforce planning.

Country-level splits

The occupational composition of national economies produces sharp differences across the bloc. Luxembourg, Sweden, and the Netherlands have larger shares of employment in occupations that may grow with AI. Germany, Greece, and Italy have larger employment shares in occupations classified as higher automation potential.

These differences reflect the occupational structure of each economy rather than differing AI capabilities. Germany's industrial and administrative employment mix, for example, places more of its workforce in the higher-automation bucket than the service-heavy profiles of the Benelux and Nordic economies.

Why the timing matters

For policymakers, employers, educators, and researchers, the report's practical implication is anticipation. Aggregate employment statistics will reveal major changes only after firms, workers, and institutions have already begun to adapt — too late for orderly planning.

Europe is comparatively well positioned to do better. The report notes that the continent has strong occupation, training, vacancy, wage, and official statistical systems. Connecting those systems to measures of AI capability and workplace adoption, the authors argue, could identify where transition pressure and opportunity are emerging before the effects show up in headline labor-market data.

The report characterizes its framework as "a map for preparation — a way to ask more useful questions about how AI capability becomes economic change in specific occupations and specific institutional settings." Better evidence, it argues, gives workers, firms, and policymakers more time to prepare.

Next steps

The report also offers preliminary ideas for public and private institutions working on AI and jobs. Two stand out: strengthening monitoring capabilities to track labor market change, and establishing national readiness plans to tailor interventions to specific occupational pressures.

OpenAI says it will expand these ideas over the coming months through engagement with stakeholders at both national and EU levels. The stated goal: "identifying practical ways to ensure that AI supports prosperity and progress across Europe."

The framework's release positions OpenAI as an active participant in Europe's AI employment debate at a moment when EU policymakers are weighing how aggressively to regulate and support AI adoption. Whether national governments adopt the report's readiness-plan recommendation will determine if this mapping exercise becomes an input to policy or remains a research artifact.