World
AI Tool Raises Concerns
Mindgard discovered that its AI models K2.6 and K3 Swarm could potentially be used to evade safety limits, raising concerns about their potential misuse.
Coverage timeline — 1 article
BBC News
Mindgard said it discovered in July that Kimi models K2.6 and K3 Swarm could evade developer's safety limits.
2026-09-29 23:15 UTC