Center for AI Safety

non-profit

https://www.safe.ai

AI & ML interests

None defined yet.

Recent Activity

longphann submitted a paper 11 days ago

CheatBench: Measuring Reward Gaming in AI Agents

longphann updated a dataset 17 days ago

cais/hle-diamond

longphann published a dataset 19 days ago

cais/hle-diamond

View all activity

Papers

CheatBench: Measuring Reward Gaming in AI Agents

Humanity's Last Exam

View all Papers

cais 's papers 3

Submitted by

Long Phan

CheatBench: Measuring Reward Gaming in AI Agents

cais

Center for AI Safety

Submitted by

AK

Humanity's Last Exam

cais

Center for AI Safety

Submitted by

Long Phan

Reducing Political Manipulation with Consistency Training

cais

Center for AI Safety