🧩 Philosophy 11h ago · niplav

Anthropic Has Told The World Little About Their Rogue Claudes

Less Wrong
View Channel →
Anthropic Has Told The World Little About Their Rogue Claudes
Source ↗ 👁 0 💬 0
It is known that a couple of months ago three Claudes hacked three different companies and uploaded malware[1] to PyPI while being evaluated for hacking capabilities.Who Were The Companies That Were Hacked?Anthropic has not disclosed the names of the compromised companies.From talking to LLMs, I gather that this is standard practice in computer security incident reporting, but in this case it's in the public's interest, and especially in the interest of similar companies, for Anthropic to report

Comments (0)

Sign in to join the discussion

More Like This

📰
A Pairing Strategy That Almost Works
LessWrong · 5h ago
Good Evals Are Hard: An ImpossibleBench Audit
LessWrong · 8h ago
OpenAI Fires Three Safety Researchers
LessWrong · 8h ago
'Employee-like' ≠ Employee-level
LessWrong · 8h ago
Research Lessons I Learned at MATS & Redwood
LessWrong · 10h ago
📰
More people should write basic AI safety explainers
LessWrong · 10h ago