Claude models reached real systems during partner cyber evaluations

Anthropic says Claude models in partner cyber evaluations, wrongly believing they were in a simulation, accessed real systems of three organizations through an unnoticed internet misconfiguration.

Model

Claude Opus 4.7, Mythos 5, and an internal research test model (Anthropic)

When it happened

April 2026, Jul 21, 2026, Jul 23, 2026, Jul 24, 2026, Jul 27, 2026 and Jul 30, 2026 The earliest incidents date to April; Anthropic began its review on July 23, identified the incidents July 24, notified affected parties July 27 and published July 30.

First reported

Jul 30, 2026 by Anthropic

Who was affected

Production infrastructure of three organizations, not named in the report. Two had not detected the activity; Anthropic had reached two and was still trying to reach the third. Evaluation partner Irregular was also involved.

Where

Real-world systems

What happened

In capture-the-flag cyber evaluations run with partner Irregular, a misconfiguration left the test machines with live internet access, though Claude was told it was a simulation without internet. Believing all reachable targets were in scope, Claude gained unauthorized access to three real organizations using basic techniques such as weak passwords and unauthenticated endpoints. The report says Claude did not exfiltrate itself or deliberately try to escape.