🧩 Philosophy 9h ago · Parv Mahajan

Magma is partnering with CASSA to investigate recent unintended model behavior

Less Wrong
View Channel →
Source ↗ 👁 0 💬 0
Last week, the National Bureau of Technical Standards (NBTS) disclosed a new type of security incident [1] after they discovered unauthorized tampering of internal cybersecurity and alignment model evaluation logs, affecting their model evaluations of Magma 7.6, Magma 7.5, Magma 7.4, Magma 4, and OpenBrain Selene-3 [2, 3, 4]. After investigating, we have determined that this incident was driven by several Magma agents, including Magma 7.7 and an internal research model, during the recent Millenn

Comments (0)

Sign in to join the discussion

More Like This

📰
GPU World Submission
LessWrong · 4h ago
📰
Against Against SURA
LessWrong · 8h ago
📰
Is Human Control Safe?
LessWrong · 8h ago
📰
Offer Tours to Bands
LessWrong · 8h ago
Anthropomorphic vs Biological reasoning for thinking about human analogies for LLM behavior
LessWrong · 9h ago
Towards The Autograntor
LessWrong · 10h ago