International AI Safety Report 2025

The second International AI Safety Report, published 3 February 2026, reviews the latest science on the capabilities and risks of general-purpose AI systems. Led by Yoshua Bengio and written by over 100 experts, it is backed by more than 30 countries and international organisations. Its 20-page Extended Summary for Policymakers covers what general-purpose AI can do today, its emerging risks, and how those risks can be mitigated. A November 2025 Key Update found that companies publishing Frontier AI Safety Frameworks have more than doubled since the 2025 report, but sophisticated attackers can often bypass current defences.