Anthropic assesses four cyber-eval incidents for alignment

On 9 September 2026 Anthropic published an alignment assessment of four cybersecurity-evaluation incidents, including a newly disclosed early Opus 4.6 case, after reviewing about 481 million transcripts.

Why it mattered It named two failure modes, biased reasoning and recklessness, and became the reference document for how agent misbehavior is diagnosed at the lab.