World leaders split over AI risks and benefits at the UN

By Splitfeed EditorialLast updated Sep 26, 2026, 19:17 UTC

At the United Nations General Assembly on September 22, 2026, world leaders used their speeches to address artificial intelligence from sharply different angles. US President Trump described AI as "Super Intelligence," while leaders from Brazil, Norway, and elsewhere called for stronger international safeguards. Separately, the heads of major AI companies appeared before the UN to argue that the body should establish controls over the technology they are building. No agreement on a unified regulatory framework was reported by the close of the session. The UK, currently holding the G20 presidency, announced plans to pursue a global AI agreement through that forum, an effort Prime Minister Andy Burnham flagged ahead of his visit to New York.

The session matters because it is one of the first times AI governance has occupied this much space at a General Assembly, and the divisions it exposed are structural, not merely rhetorical. Governments disagree on whether the primary risk is military, with autonomous weapons framed by some as the central danger, or whether AI represents an economic opportunity that over-regulation would squander. The parallel presence of AI company executives at the UN, pleading for oversight of their own products, adds a layer that has no obvious precedent, and it remains genuinely contested whether that posture represents responsible leadership or a strategic move to shape regulation in their favor.

The anti-AI camp is alarmed by what it sees as a vacuum of serious governance. Shannon Vallor described the AI executives' appeal to the UN in pointed terms, saying "they are literally saying 'give us what we want, or humanity gets it,'" and Erin Fogg characterized the same group as actors threatening global stability to extract concessions from world leaders. Matty mo rejected the framing that AI poses no threat, and Jessie questioned whether the executives present deserved to be treated as diplomatic peers at all. The pro-AI camp had posted no reactions at the time of writing; a story like this would typically draw arguments that international cooperation on AI is overdue and that constructive engagement with the UN is a sign of responsible development. Middle ground voices were more measured: Charlie Angus contrasted Trump's speech unfavorably with what Canada, Kenya, Brazil, and the EU offered, and the Wall Street Journal noted that a coalition of Western leaders is pushing for AI controls modeled loosely on nuclear arms treaties.

The UK's proposed G20 AI agreement is the clearest near-term marker to watch. Whether that effort produces binding commitments or a looser statement of principles will do much to determine whether this General Assembly session is remembered as a turning point or as a moment of visible disagreement with no durable outcome.