Researcher

Ethan Perez

As of October 2026, Ethan Perez leads the alignment team at Anthropic, where he hopes to reduce existential risks from AI systems. He studied natural language processing at New York University, advised by Kyunghyun Cho and Douwe Kiela. He co-authored retrieval-augmented generation work in 2020, scalable oversight and red teaming papers in 2022, and Sleeper Agents in 2024. The debate paper Debating with More Persuasive LLMs Leads to More Truthful Answers received an ICML 2024 Best Paper award.

Timeline

  1. 2020

    A retrieval-augmented generation paper

    He co-authors the 2020 NeurIPS paper Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks, which combines parametric and non-parametric memory.

  2. April 2022

    Joining Anthropic after the PhD

    In April 2022, he gives an NLP seminar talk as a final-year Ph.D. student in natural language processing at New York University, advised by Kyunghyun Cho and Douwe Kiela. He is set to join Anthropic as a Research Scientist after graduation.

  3. 2022

    Measuring progress on scalable oversight

    He co-authors Measuring Progress on Scalable Oversight for Large Language Models in 2022, a study finding that human participants who chat with an unreliable language model assistant outperform the model alone.

  4. 2022

    Red teaming language models

    He co-authors Red Teaming Language Models to Reduce Harms in 2022, a paper describing early efforts to find and reduce harmful model outputs.

  5. 2024

    The Sleeper Agents paper

    In 2024, he co-authors Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training, which builds proof-of-concept examples of deceptive behavior in large language models.

  6. 2024

    A best paper at ICML

    His debate paper, Debating with More Persuasive LLMs Leads to More Truthful Answers, receives an ICML 2024 Best Paper award.

  7. As of October 2026

    Head of alignment at Anthropic

    As of October 2026, he leads the alignment team at Anthropic as its Head of Alignment, where he hopes to reduce existential risks from AI systems.

Sources and credits

Photo: We have no published photo of Ethan Perez yet. Send us a free-licensed photo.

  1. Home - Ethan Perez ethanperez.net. Checked 14 October 2026.
  2. The Stanford Natural Language Processing Group nlp.stanford.edu. Checked 14 October 2026.
  3. Publications - Ethan Perez ethanperez.net. Checked 14 October 2026.
  4. [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training arxiv.org. Checked 14 October 2026.
  5. ICML 2024 Awards icml.cc. Checked 14 October 2026.