UK AI Security Institute Retains Early Access to Frontier US Models
Britain has secured continued early access to test advanced American artificial intelligence models for pre-release safety evaluations.
FTMQ SI, written by our newsroom0 views
Britain's Artificial Intelligence Security Institute has secured continued pre-release access to evaluate advanced frontier artificial intelligence models developed in the United States. AI Minister Kanishka Narayan confirmed the agreement on October 12, 2026, according to BigGo Finance. The announcement follows a decision by the Trump administration to lift restrictions that previously blocked American companies from sharing their newest models with the British safety body, ABAB News reported. [1][2]
Scientists at the institute are currently conducting evaluations on unreleased systems, including Google's Gemini 4 Argon, BigGo Finance reported. The system has not been released to the public due to unresolved safety concerns. The safety tests proceed alongside warnings from the British government that a major cyber breach remains likely or inevitable. [1]
The confirmation resolves weeks of uncertainty regarding access for British testing teams. According to BigGo Finance, earlier reports indicated that the White House had requested artificial intelligence developers, including OpenAI and Anthropic, to withhold their latest models from British testers pending administrative reviews. [1]
In short
- Britain's AI Security Institute maintained pre-release safety testing access for frontier US models.
- UK AI Minister Kanishka Narayan confirmed scientists are testing Google's unreleased Gemini 4 Argon model.
- The Trump administration lifted previous restrictions on US companies sharing AI models with the UK institute.
Sources
Every paragraph above points to the numbered items it rests on. Read the originals here.
- [1]UK Confirms Early Access to Test Latest US AI Models, Cites Security RisksBigGo Finance, 11h ago (the report this story comes from)
- [2]Trump Administration Lifts Restrictions on US Companies Sharing AI Models with UK Security Institutehttps://www.ababnews.com/, 16h ago
Background
- [3]Mistral AI on Wikipedia
- [4]Paramount+ on Wikipedia
- [5]Risk on Wikipedia
Our newsroom writes these reports with the help of software, from the 5 sources listed and nothing else, and checks them against those sources. Facts can still be wrong or move on; the originals are the record. Spotted a mistake? Write to daniel@monsterkong.com.
Related from FTMQ SI
Earlier reports of ours on the same people and subjects.
- Anthropic cuts off internet access for internal AI evaluationsTop Stories, 1d ago
- Google Announces Gemini 4 Argon Model for Enterprise and Cybersecurity WorkflowsResearch Papers, 7d ago
- Anthropic launches free automated security scanner for open source projectsTop Stories, 4d ago
- Microsoft chief Satya Nadella urges emergency brakes for advanced artificial intelligenceTop Stories, 2d ago
- Engineers test AI models behind the wheel of a Toyota CorollaTop Stories, 5d ago
More in Industry Labs
Get the day in one email
Reports like this one, the top news of the last 24 hours, every morning. Free, one email a day; readers can comment under every report.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.
Comments
Loading
Join the conversation
Comments are open to readers of our daily email: the top news of the last 24 hours, every morning, free. Sign up and the comment box opens.
Already on the list? Enter the same address and we will send a sign-in link.
By signing up you agree to our terms and privacy policy. Unsubscribe any time.