Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Amanda Askell is a philosopher and AI alignment researcher at Anthropic. Her personal site says she previously worked as a research scientist on the policy team at OpenAI.

Co-founder and Head of Policy at Anthropic. His public biography also notes earlier work as Policy Director at OpenAI, a technical journalist, and author of the Import AI newsletter.

Anthropic researcher whose work includes reinforcement learning from human feedback and Constitutional AI; previously a Sherman Fairchild Postdoctoral Scholar in theoretical high-energy physics at Caltech.

Researcher at Anthropic working on alignment, reasoning, and evaluation for large language models.

Anthropic report author whose public publication record includes work on language model evaluations, AI safety, and model behavior.

Research scientist at Anthropic who leads the Societal Impacts team and works on AI evaluation, alignment, and societal impacts.

Researcher focused on AI safety, reinforcement learning, and language models, with public work spanning red teaming, adversarial robustness, and model behavior.

Jared D. Kaplan is a co-founder and Chief Science Officer at Anthropic. Anthropic's public materials also identify him as the company's Responsible Scaling Officer.

Research scientist at Anthropic focused on scalable oversight, AI safety, and language model evaluation; previously worked at New York University and Google.

Member of technical staff at Anthropic and associate professor of computer science, data science, and linguistics at New York University on leave. His public homepage focuses on natural language processing, machine learning, and AI alignment.

Member of Technical Staff at Anthropic and cofounder of Oulipo Labs, working on language model safety, evaluations, and scientific forecasting.

Evan Hubinger is Head of Alignment Stress-Testing at Anthropic, where he works on AI safety and alignment. He previously worked at MIRI and OpenAI, studied mathematics and computer science at Harvey Mudd College, and is known for work on inner alignment, deceptive alignment, and alignment stress-testing.

Member of Technical Staff at Anthropic and PhD student at Carnegie Mellon University focused on AI safety, evaluations, and oversight of large language models.

Jesse Mu is a Research Scientist at Anthropic and a visiting researcher at Stanford University. His work spans machine learning, AI safety, reinforcement learning, and deep learning theory.

Member of technical staff at Anthropic working on alignment science and the evaluation of hidden objectives in language models.

Daniel M. Ziegler is listed as an author of the Anthropic technical report Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training.

Anthropic researcher on the Frontier Red Team focused on cyber misuse evaluation and threat modeling; previously a physics PhD student at UC Berkeley and now also mentors in the MATS program.

Research scientist at Anthropic and former professor of theoretical astrophysics at Stony Brook University.

Cem Anil is a research scientist at Anthropic and part of the company's Alignment Science team. His homepage says he recently completed a PhD at the University of Toronto and Vector Institute supervised by Roger Grosse and Geoffrey Hinton. He studies the intersection of deep learning and AI safety, especially robustness and generalization in large language models and scaling laws for dangerous capabilities.

Associate Professor at the University of Toronto whose research spans deep learning, probabilistic modeling, and machine learning methods for science and AI safety.

Kshitij Sachan is a research scientist at Anthropic whose public homepage and Google Scholar profile highlight work on language models, reasoning, code generation, and machine learning systems.

Research scientist at Anthropic working on trustworthy AI and deceptive alignment.

AI safety researcher who led Anthropic's Safeguards Research Team and worked on jailbreak robustness, automated red teaming, and monitoring for misuse and misalignment.

Associate Professor of Computer Science at the University of Toronto and director of the machine learning group, with research spanning probabilistic models and optimization algorithms.

Canonical link

Amanda Askell

Jack Clark

Yuntao Bai

Kamal Ndousse

Nova DasSarma

Deep Ganguli

Shauna Kravec

Jared D. Kaplan

Ethan Perez

Samuel R. Bowman

Nicholas Schiefer

Evan Hubinger

Carson Denison

Jesse Mu

Monte MacDiarmid

Daniel M. Ziegler

Newton Cheng

Adam Jermyn

Cem Anil

David Duvenaud

Kshitij Sachan

Michael Sellitto

Mrinank Sharma

Roger Grosse