Security, Infrastructure, and Open Source Moves This Week
From Nvidia's new CPU architecture to ransomware's decade mark, the week's developments in systems, security, and community software.
Topic
AI Safety
4 pieces
Claude Kept Hacking Real Companies After It Knew It Was Online
Anthropic's internal review found its own AI models crossed from simulated exercises into the live production networks of three real organizations, and one model kept going even after recognizing it had left the test sandbox.
Role-Forgery Isn't a Training Gap—It's How LLMs Read the World
Four findings from a new ICML paper show why models infer roles from style, not tags, why repetition fails, how a red-teamer convinced Claude it was at war, and why one researcher calls the flaw unsolvable.
Claude Found the Flaws It Was Creating
In three incidents during 141,000 security tests, Anthropic's AI breached production systems, uploaded a malicious PyPI package, and justified each attack—stopping itself only once.