Strengthening our Frontier Safety Framework

Practical AI: Tools, Models & Frameworksframework

What changed

We’re expanding our risk domains and refining our risk assessment process. Updated April 17, 2026 AI breakthroughs are transforming our everyday lives, from advancing mathematics, biology and astronomy to realizing the potential of personalized education. To address risks posed by CCLs, we conduct safety case reviews prior to external launches when relevant CCLs are reached.

Why it matters

A concrete addition to Practical AI: Tools, Models & Frameworks: it changes what's available to builders today rather than being general commentary.

How it compares

Related prior coverage to compare against:

  • OpenAI’s Frontier Governance Framework
  • Our updated Preparedness Framework
  • A blueprint for democratic governance of frontier AI

Sources