C
Author
Center for AI Safety
By and about Center for AI Safety
news
Fable 5.2 tops Epoch and Remote Labor Index, but not every test
Independent tests put Claude Fable 5.2 first on Epoch's Capabilities Index and the Remote Labor Index, where it automates 41.3% of real freelance projects, while it trails on some games and legal tasks.
newsAI 2027 is running early: METR’s Anthropic report resets the clock
METR estimates AI has roughly doubled Anthropic’s pace of progress, and AI 2027’s authors now expect a superhuman AI researcher in 2027; skeptics point to the cost curve.
newsEigenism: Hendrycks wants AI safety to engineer identity, not cages
Dan Hendrycks proposes eigenism: identity as a graded, distributed pattern, and outcomes scored as everyone’s wellbeing weighted by how much of your pattern they carry. Its safety upshot is that personalization and privacy become alignment tools.