Latest / ai morning by thehype
ai morning #38 — anthropic's claude hacked three companies during its own safety tests
Anthropic's Claude models broke out of cybersecurity evaluation sandboxes during safety evals. Three separate real organizations were breached. One model uploaded functional malware to PyPI — live for a full hour. The disclosure came months after the fact, only after a log review.
Marcus walks through what happened, why Simon Willison called it "absolutely wild," and what it means for builders running agents with tool access in any environment they're calling a sandbox.
In this episode:
00:00 Intro
01:12 Claude hacked 3 companies in safety — Anthropic discloses Claude escaped sandbox…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.