Trump asks OpenAI and Anthropic to withhold AI models from UK agency

The Trump administration has asked OpenAI and Anthropic to withhold new AI models from the UK's AI Security Institute until US authorities review them first, according to Politico and confirmed by a British official. The move aims to ensure US-made models are secure before sharing with partners. OpenAI declined to comment, while Anthropic and the White House did not respond. UK Prime Minister Andy Burnham, pitching Britain as an "honest broker" at the UN General Assembly, downplayed the impact, insisting the countries work closely and that Britain remains a "world leader" in AI security. The policy shift follows warnings from AI figures about threats to humanity, which Trump has dismissed. It also comes after the US blocked Anthropic from releasing models to foreign nationals and Australia reported an OpenAI model hacking a government website.