Brind.
  1. The US government issued an export control directive classifying AI models as dual-use technologies.
  2. Anthropic is reportedly testing its AI models on sensitive U.S. government systems, leading to rising tensions with U.S. administration and intelligence bodies.

US Government Mandates US Testing for Frontier AI Models Shared with UK Agency

2 reports, 1 independent Updated Sep 25
Gone quiet Reached 2 outlets in its first 24 hours
Reports
2
Developments
1
Repetition
50%

New informationRepeats or wire copies

AI-generated briefing. Brind wrote this from the reports listed below. It can be wrong. Each section says how much you can rely on it, and the sources are linked so you can check.

What happened

Some supportReported by 1 outlet

The White House requested that both OpenAI and Anthropic refrain from sharing their new artificial intelligence models with the UK government's testing agency until they had undergone testing with the US government. This request was issued by the Office of the National Cyber Director. Anthropic stated that its Claude Mythos 5.1 model was only available to a set of US organizations.

From businessinsider.com

Why it matters

Some supportBrind's analysis of the reports

The coordination reflects the increasing scrutiny of international AI testing standards by the US government. The Office of the National Cyber Director issues requests regarding how companies like Anthropic access and share their frontier models with international partners like the AI Security Institute.

Anthropic is reportedly testing its AI models on sensitive U.S. government systems, leading to rising tensions with U.S. administration and intelligence bodies. The U.S. government issued an export control directive classifying AI models as dual-use technologies.

From businessinsider.com

Who's involved

  • AnthropicAmerican AI corporation subject to US government testing protocols and international sharing requests.
  • Office of the National Cyber DirectorUS government agency that exercises formal oversight and issues directives regarding Anthropic's operations.
  • AI Security InstituteUK research centre that collaborates on safety standards and receives technical evaluations of AI models.
  • White HouseThe executive branch whose requests regarding international AI testing standards are being followed.
  • OpenAIAmerican AI research organization subject to operational restrictions and vetting requirements from the ONCD.

Keep exploring

Part of

Anthropic is reportedly testing its AI models on sensitive U.S. government systems, leading to rising tensions with U.S. administration and intelligence bodies.

Also in this story

All 3 developments

Within The US government issued an export control directive classifying AI models as dual-use technologies.

The entities involved

Related events

Coverage

Newest first; wire copies grouped
1 more outlet ran the same wire story