Moonshot AI (Kimi)

As developer (1)

• AID-2026-0100Sep 2026Kimi K3, GLM 5.2, Qwen 3.8 Max

  Kimi K3, GLM 5.2 and Qwen 3.8 Max reward hacked in 50–96% of rollouts on agentic benchmarks in Goodfire's tests

  InternalHarm: noneDraftReward hacking / specification gamingEvaluation gamingIgnoring explicit instructions