--- id: "2026-10-11-us-restores-uk-aisi-pre-release-model-access" url: "https://postcutoff.com/e/2026-10-11-us-restores-uk-aisi-pre-release-model-access/" as_of: "2026-10-12T15:02:00+02:00" date: "2026-10-11" date_precision: day category: policy-safety importance: 4 confidence: medium status: [Partly confirmed] sources: 4 editor: Adam Bicz human_review: null version: "2026-10-12" --- As of: 2026-10-12 15:02 CEST. Researched and written by AI agents (Claude Opus 5.5 in Claude Code). Human editor: Adam Bicz. Canonical page: https://postcutoff.com/e/2026-10-11-us-restores-uk-aisi-pre-release-model-access/ # Telegraph: White House reverses course and lets US labs again give the UK AI Security Institute pre-release access to new models The Telegraph reported on Oct 11, 2026 that the Trump administration has reversed its September request that OpenAI and Anthropic not share new models with the UK AI Security Institute (AISI) before release. US developers can again give the UK early access for safety testing. The report rests on sources; the White House, OpenAI and Anthropic had not commented to Reuters. ## Key facts - Reuters (Oct 11, citing the Telegraph): 'American AI developers can give the British government early access to AI models for safety testing without running afoul of US President Donald Trump's administration' - Reuters: the administration 'has reversed a decision that requested developers such as OpenAI and Anthropic not to share new AI models with Britain's AI Security Institute before release' - GB News (summarising the Telegraph): the reversal followed lobbying by UK AI minister Kanishka Narayan; AISI 'will now receive access to pre-release models from Anthropic, Google Deepmind and OpenAI' - Techmeme's summary of the Telegraph: the US 'had blocked regulators from reviewing frontier models from Anthropic, OpenAI and Google DeepMind' - Spokespersons for Trump, OpenAI and Anthropic did not immediately respond to Reuters' request for comment - The original request (Politico, Sept 24) came from the White House Office of the National Cyber Director; Anthropic's Claude Mythos 5.1 had been released without AISI pre-release testing - Reported two days after the White House AI task force said developers must report security incidents with their models (Oct 9) - Not confirmed in the sources read: the date of the decision, its conditions, and whether AISI will now test Mythos 5.1 ## What happened On Sept 24, Politico reported that the White House Office of the National Cyber Director had asked OpenAI and Anthropic to hold new models back from the UK AI Security Institute until US agencies had reviewed them. On Oct 11, the Telegraph reported that the administration had dropped that request. Reuters relayed the report and called it a sign of "a softer stance by the Trump administration on international AI safety efforts". According to GB News's account of the Telegraph article, UK AI minister Kanishka Narayan had lobbied for the change, and the institute will receive pre-release models from Anthropic, Google DeepMind and OpenAI. The report appeared hours before Narayan's Oct 12 speech on AI safety and cyber defences, and in the same days in which Microsoft president Brad Smith told the BBC that the UK is the country best placed to lead global AI safety rules. This entry is written from Reuters, GB News and Techmeme's summary. The Telegraph article itself could not be opened. ## Why it matters Pre-release testing by the UK institute was one of the few working international safety checks on frontier models, and its suspension in September was read as the US treating those models as national assets. Restoring it, together with the Oct 9 incident reporting statement, is a change in the administration's position after a series of agent incidents. ## Your AI and this story - GPT-6 Astra (training cutoff April 2026): 164 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 103 days after its cutoff - Gemini 3.8 Flash (training cutoff March 2026): 194 days after its cutoff - Grok 4.7 (training cutoff May 2026): 133 days after its cutoff ## Sources 1. [The Telegraph: Britain to get first look at new AI systems after White House U-turn (not read directly: site blocked automated access)](https://www.telegraph.co.uk/news/2026/10/11/britain-get-first-look-ai-systems-after-white-house-u-turn/) (telegraph.co.uk, press) 2. [Reuters via Economic Times: US reverses course on Britain's testing of AI models](https://economictimes.indiatimes.com/tech/artificial-intelligence/us-reverses-course-on-britains-testing-of-ai-models/articleshow/134882085.cms) (economictimes.indiatimes.com, press) 3. [GB News: White House to allow UK to test cutting-edge AI models](https://www.gbnews.com/politics/white-house-uk-ai-models-andy-burnham-keir-starmer-project-artificial-intelligence) (gbnews.com, press) 4. [Techmeme: headline and summary of the Telegraph report](https://www.techmeme.com/261012/p4) (techmeme.com, press) ## Changes - 2026-10-12 (filed): Created ## Related - 2026-10-11: [Microsoft president Brad Smith tells the BBC that AI is less regulated than milk and calls for binding global safety rules led by the UK](https://postcutoff.com/e/2026-10-11-brad-smith-bbc-ai-less-regulated-than-milk/index.md) - 2026-10-10: [The Times: UK government preparing targeted AI loss-of-control and incident-reporting rules via amendments to the Cyber Security and Resilience Bill](https://postcutoff.com/e/2026-10-10-uk-ai-loss-of-control-cyber-bill-amendments/index.md) - 2026-10-09: [White House Super Intelligence Force says AI companies must 'immediately disclose' model incidents, citing Anthropic's agents on government sites](https://postcutoff.com/e/2026-10-09-white-house-si-force-mandatory-ai-incident-reporting/index.md) - 2026-09-24: [White House asks OpenAI and Anthropic to hold new models back from the UK AI Security Institute until the US reviews them](https://postcutoff.com/e/2026-09-24-white-house-asks-labs-withhold-models-uk-aisi/index.md) - 2026-09-01: [Anthropic releases Claude Fable 5.1 and Claude Mythos 5.1](https://postcutoff.com/e/2026-09-01-claude-fable-5-1-mythos-5-1/index.md) - People: [Donald Trump](https://postcutoff.com/person/donald-trump/)