US Government to Implement Pre-Release Vetting for Major AI Models

Key Intelligence

• ■Google, Microsoft, and xAI, among other major AI developers including OpenAI and Anthropic, have agreed to provide the US government with early access to their frontier AI models for national security and safety reviews.
• ■The White House, under the current administration, is actively considering an executive order or new policies to mandate pre-release vetting for high-risk AI models.
• ■The Commerce Department's AI Safety Institute (CAISI) and other federal agencies are expanding their capabilities to conduct comprehensive evaluations of these advanced AI systems.
• ■This initiative signifies a shift towards proactive federal oversight, aiming to identify and mitigate potential national security and public safety risks before powerful AI technologies are widely released.