How AI's acceleration created a global policy vacuum, as EU AI Act enforcement lags and regulators remain torn between harnessing AI and fearing its risks
Ursula von der Leyen, the president of the European Commission, gathered senior officials for an exclusive meeting at a 19th-century palace outside Brussels this month.
Context & Ripple Effects
The EU’s 2023 political deal on a comprehensive AI rulebook was meant to set limits for sensitive uses, but businesses were already warning in 2024 that key implementation details were missing. A 2025 postponement of a key component sharpened the critique that the bloc’s regulatory ambitions were outpacing execution.
The Commission has also framed AI as an investment and competitiveness priority, including von der Leyen’s €200 billion AI investment plan announced in Paris. The officials’ meeting puts that unresolved tension—building AI capacity while making the Act enforceable—at the center of EU policy.
First-order effects
- The European Commission must reconcile delayed AI Act enforcement with its push for AI adoption, leaving affected businesses without the implementation clarity critics had sought since the law took effect.
- Von der Leyen’s senior-policy team faces a more immediate trade-off between rules designed for high-risk AI uses and an industrial agenda intended to narrow Europe’s AI gap with the US and China.
Second-order effects
- AI developers and companies deploying AI in Europe must plan around an unsettled compliance timetable, weakening the predictability that a comprehensive rulebook was intended to provide.
- The delay strengthens the argument, advanced after the postponement of a key AI Act measure, that EU AI policy must pair enforceable obligations with a more credible path to innovation and deployment.
Third-order effects
- If implementation continues to trail technical and commercial change, the EU’s influence will depend less on passing broad AI rules than on whether it can turn them into operational standards that companies can follow.
- The policy model is shifting toward joint governance and industrial policy: regulatory legitimacy increasingly rests on whether public institutions can support domestic AI capacity while policing harmful uses.
The trend: AI governance is becoming inseparable from AI industrial policy as governments seek rules that can keep pace with deployment without sacrificing competitiveness.