The U.K. AI Security Institute and the U.S. Center for AI Standards and Innovation publish a joint evaluation of Kimi K3 days before its weights are due. They find its safeguards did not stop it attempting cyber-exploit development or offensive cyber operations.
Why it's scored this way
Government testers confirmed the weak-safeguards worry on the leading open model.
Sources
- What the UK AISI Kimi K3 Evaluation Means for Enterprise AI Security, Zerberus, Jul 2026