Taking the Rulebook to Manhattan: McGrath Pitches EU AI Vision at UN

Speaking at a United Nations General Assembly side-event in New York on 28 September 2026, European Commissioner Michael McGrath called for international guardrails to ensure democratic societies retain control over artificial intelligence. Addressing diplomats and policy experts at a panel on 'Shaping Democratic Agency over the course of AI Development', McGrath argued that democratic institutions must actively steer technology rather than passively react to algorithmic shifts. The Commission highlighted how public oversight, algorithmic transparency, and human supervisory requirements are essential to safeguard democratic processes, electoral integrity, and fundamental rights against potential disruption from high-risk automated systems.

For citizens, the debate over democratic agency in AI determines how much control human institutions retain over automated systems affecting daily life. Algorithms increasingly influence news feeds, administrative decisions, credit assessments, and election campaigns. Without international standards, automated tools risk entrenching bias, spreading coordinated disinformation, or mishandling personal data without explicit public oversight. Establishing common democratic guardrails aims to ensure that citizens and accountable governments—rather than private tech platforms or authoritarian states—set the rules for how artificial intelligence operates.

The New York address reflects a classic Brussels tradition: taking European regulatory frameworks on an international tour in the hope that global partners will adopt them. Having negotiated complex internal digital laws, the European Commission routinely converts its domestic legislative rulebooks into diplomatic soft power. UN side-events—informal discussions staged alongside the formal diplomatic summit—offer the ideal stage for this regulatory evangelism. Behind the polished speeches lies a pragmatic attempt by Brussels to export its standards globally before competing economic blocs or tech conglomerates establish their own norms as the default.

Following the UN General Assembly side-event, international working groups will continue drafting multilateral benchmarks for trustworthy AI and democratic protections. Inside the EU, the Commission remains focused on implementing its existing digital governance framework, building oversight bodies for high-risk AI models, and coordinating with international partners to align technical safety standards ahead of upcoming bilateral and global tech forums.