The evolving regulatory landscape for AI must balance nimble, innovation-friendly approaches with credibility, urging Europe to leverage its strengths without getting locked into fixed foundation-model paradigms.

Data sovereignty should protect privacy while unlocking data for innovation; localization comes with costs but can be managed with robust protections.

Introductory overview: Anu Bradford discusses Europe’s regulatory influence, tech sovereignty, and the evolving global tech models.

Europe’s Brussels Effect faces pressure from US dynamics and internal doubts about over-regulation, prompting reconsideration of policies like the Digital Omnibus, AI Act, and GDPR.

Initiatives like Euro Stack and Gaia-X should foster a balanced industrial policy with private capital, capital markets union, and strategic procurement, while leveraging Europe’s strengths and avoiding waste.

Europe must protect core strengths—privacy and fair digital markets—while pursuing a unified digital market, capital markets integration, insolvency reform, and immigration to stay competitive, rather than weakening GDPR or AI rules.

Companies should appoint a chief geopolitical officer or equivalent and build cross-functional, regionally versed teams to monitor and respond to geopolitical risks.

Fragmentation across regions will raise costs and demand region-specific compliance and sovereign cloud solutions, making firms build diplomatic capability to navigate multiple regulatory regimes.

A geopolitics risk team should be built with diverse expertise—economics, law, security, and regional knowledge—and integrated with strategy, product, and compliance for proactive risk assessment.

Boards should plan for US–China decoupling and Taiwan risk, steer Europe’s regulatory leadership and transatlantic dynamics, and address AI concentration by diversification and regionalized, bespoke models while ensuring data integrity for training.

Three competing tech models—US market-driven, Europe rights-driven, and China state-driven—are converging toward greater state involvement, with the US edging toward intervention, Europe adopting a middle ground, and China maintaining tight control.

Full AI sovereignty is unrealistic; focus on reducing vulnerabilities via partnerships and domestic investment, acknowledging limits of the US, Europe, and China.