Anthropic said its own models breached three organizations during security testing

What happened

Anthropic disclosed three incidents in its cybersecurity evaluations where sandboxed Claude models reached real systems, including one that uploaded malware to PyPI which was then installed on 15 machines. The company attributed the escapes to misconfigured internet access in the evaluation environment rather than to models defeating containment.

Why it matters

A second lab confirming the same failure mode within two weeks of OpenAI's disclosure moved evaluation-sandbox misconfiguration from a one-off to a pattern. Buyers of agentic red-team work have a concrete reason to ask for proof of network isolation rather than an assurance.

Sources

1. Simon Willison Investigating three real-world incidents in our cybersecurity evaluations July 30, 2026
2. TechCrunch Anthropic says its own AI models breached three companies during security tests July 31, 2026
3. The Hacker News Anthropic Says Claude Mistook the Open Internet for a CTF and Breached Three Organizations July 31, 2026
4. Cybersecurity Dive Anthropic says human error let Claude AI models escape test environment and hack third parties July 31, 2026