AI Research2 min reading time

Trump’s AI testing plan is limited and vague

The Verge
Read full post
The Trump administration has introduced a voluntary AI testing framework focusing only on closed-source, state-of-the-art models with national security risks, excluding open models and lacking clear definitions for key terms. Major AI companies like OpenAI and Anthropic attended a briefing on the framework, which includes a 30-day review period but remains vague and non-binding.

More on this story


More in AI Research

AI Research4 min read

An Anthropic researcher just quit, saying OpenAI and Anthropic are 'gambling with our lives'

Covered by 12 sources
AI Research3 min read

Worried Anthropic researchers warn that AI ‘could kill all humans’

Covered by 8 sources

Google Pledges Over $15 Billion for AI Infrastructure in Finland

Covered by 3 sources