Trump administration to vet AI models before release
SkimNews Take
Pre-release vetting by an executive working group repositions the federal government as a default gatekeeper for commercial AI deployment, even under a nominally deregulatory administration, making the "hands-off" label harder to sustain as export-control and consumer-safety functions converge in a single approval pipeline.
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Trump administration is drafting an executive order to establish an AI oversight working group that would review AI models before they are publicly released.
- AI oversight working group would be tasked with vetting high‑risk AI models and assessing security‑related risks, according to the administration’s discussion.
- Anthropic Mythos is cited as a trigger for the White House to consider pre‑release reviews of high‑risk AI systems.
- White House officials have briefed leading AI firms—including Anthropic, Google, and OpenAI—on the proposed government review process.
- The administration’s stance marks a shift from its earlier non‑interventionist approach to AI, moving toward tighter regulation and safety testing.
Why it matters: The policy gives the U.S. government authority to block AI models deemed high‑risk, directly affecting developers such as Anthropic, Google, and OpenAI by adding mandatory safety reviews, while aiming to curb security threats from advanced AI in the near term.
