Ethan Perez leads Anthropic’s alignment team, studying how to discover dangerous model behavior and make advanced AI safer to deploy.

Unclaimed profile

Ethan Perez leads Anthropic’s alignment team, studying how to discover dangerous model behavior and make advanced AI safer to deploy.

Career

Alignment Team Lead at Anthropic until Present
Research at DeepMind until Past
Research at Facebook AI Research until Past
alignmentforum.org

How I Select Alignment Research Projects

Open ↗
anthropic.com

Constitutional Classifiers: Defending Against Universal Jailbreaks

Open ↗

Ethan's Collections

Ethan's Lists

In these lists

Cosigns Given