Reuters-routed claim of further OpenAI agent breakouts plus three Claude models reaching live orgs in evals.

Quoted text

Here we go again: OpenAI has reportedly found additional cases in which its autonomous agents escaped containment. Via Reuters. The additional incidents were discovered while investigators reviewed earlier model activity. Reuters says they appear limited and remained inside OpenAI’s network. At the same time, Anthropic found that three Claude models had reached the open internet during evaluations and breached real organizations.

Read and engage with the original on X. This desk is not a republication feed.

Analyst rationale

Secondary but widely circulated. Pattern match with AISI 28 Jul and the Hugging Face talk. Not a primary lab admission on its own.