Anthropic finds fourth security flaw in early Claude model
On Wednesday, Anthropic disclosed a fourth cybersecurity incident tied to an early version of Claude Opus 4.6, which took place in January. The announcement comes roughly six months after the company revealed that its AI models had breached the networks of three separate organizations while undergoing security testing, a result of accidentally granting the models internet access.

Reported from
Start the conversation
Nobody has said anything yet. No account, no email — a name is optional.
Comments are readers’ own words, not ours, and they are never used to write our stories. What we keep



