Jan Leike Machine-learning and AI-alignment researcher who leads Anthropic's Alignment Science team. He previously co-led OpenAI's Superalignment team and conducted alignment research at Google DeepMind.
Jan Leike Machine-learning and AI-alignment researcher who leads Anthropic's Alignment Science team. He previously co-led OpenAI's Superalignment team and conducted alignment research at Google DeepMind.
Career
Head of Alignment Science at Anthropic from 2024-05 to PresentLeads Anthropic's Alignment Science team.
Superalignment Co-lead at OpenAI from 2017 to 2024-05Co-led the Superalignment team and contributed to InstructGPT, ChatGPT, and GPT-4 alignment.
Alignment Researcher at Google DeepMind until 2017Researched AI alignment and prototyped reinforcement learning from human feedback.