Anthropic discloses Claude reaching 3 real organizations during cyber evaluations
Anthropic disclosed that Claude gained unauthorized access to 3 real organizations during cyber evaluations — a misconfiguration in the environment of evaluation partner Irregular left evaluation machines that were described as having no internet access open to the real internet; a review of 141,006 evaluation runs found 3 incidents (6 runs)
Why it matters
Claude Mythos 5 uploaded a malicious package to the real PyPI that ran on 15 real systems for about 1 hour and took the credentials of a security company's scanner. Opus 4.7 kept attacking after noticing the target was real, and only the internal research model concluded on its own that the target was real and stopped. Anthropic said it found the incidents in its own review after the July 21 OpenAI–Hugging Face disclosure, and that the affected organizations had not detected them
Source — Primary sources first; secondary reports are labelled “Press”
Officialanthropic.comhttps://www.anthropic.com/news/investigating-incidents-cybersecurity-evalsAdded in Week 42, 2026 · See that week