Trump administration considering safety review for new AI models
Summary
The Trump administration is considering a plan to have the Pentagon test the safety of AI models before they are used by federal, state, and local governments. This move follows concerns about the security risks of advanced AI technologies, like Anthropic’s Mythos and OpenAI’s GPT 5.5.Key Facts
- The White House’s Office of the National Cyber Director held meetings with tech companies to discuss AI security concerns.
- The proposed plan would require the Pentagon to lead safety testing for AI models used by government agencies.
- This plan was in development before the release of Anthropic’s Mythos, an AI model that raised new cybersecurity worries.
- The Trump administration may issue an executive order to direct multiple agencies to test AI safety.
- Discussions about an AI safety framework include updating policies to address new risks from advanced AI models.
- The White House is also considering allowing government use of Anthropic’s Mythos despite a current ban.
- The plan revisits ideas from the Biden administration’s AI order, which President Trump revoked on his first day in office.
- Experts have mixed views: some worry about slowing AI use, others about possible AI-related cyberattacks.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.