Jan Leike

Anthropic alignment researcher whose personal site says he leads the Alignment Science team; previously co-led OpenAI's Superalignment team and earlier worked on reinforcement learning from human feedback at DeepMind.