Alignment and Safety | Field

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.

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

Tom Henighan works on large language model interpretability at Anthropic. He previously worked on scaling laws at OpenAI and machine learning engineering at Beehive AI, and he studied physics at Stanford after graduating from Ohio State University in 2010 with a degree in English, mathematics, and philosophy.

Aengus Lynch is a fifth-year PhD student in machine learning at Carnegie Mellon University advised by Zico Kolter. His homepage says he works on control, reinforcement learning, games, and machine learning, and his CV shows research internships at Anthropic and Google DeepMind after completing a BASc in engineering physics at the University of British Columbia.

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.

Jacob Hilton is a researcher and executive director at Alignment Research Center, where he works on mechanistic approaches to outperforming random sampling. He previously worked at OpenAI on truthfulness, reinforcement learning, and interpretability for language models, earlier worked at Jane Street, completed a PhD in mathematics at the University of Leeds, and later coauthored Anthropic work on constitutional classifiers.

Sandipan Kundu is a member of technical staff at Anthropic. His public The Org profile says he previously held postdoctoral positions at Johns Hopkins and Cornell, worked and studied at the University of Texas at Austin, and earned a master's degree in physics from the Indian Institute of Technology Kanpur.

William Saunders is a research scientist at Anthropic working on aligning and evaluating language models. His public homepage says he works at the intersection of game theory, optimization, and deep learning, previously interned at OpenAI, DeepMind, and Mila, studied mathematics at the University of Oxford, and is a PhD student in machine learning at Carnegie Mellon 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.

Researcher at Anthropic whose public report authorships and scholarly profiles cover language model evaluation, AI safety, and robustness.

Ryan Greenblatt is chief scientist at Redwood Research. His public Redwood and Forethought profiles identify him as part of Redwood's AI safety team and say he holds a BS in applied mathematics and computer science from Brown University.

Research scientist at Anthropic whose public work includes AI alignment, reinforcement learning from human feedback, and model behavior.

Senior research engineer at Anthropic interested in agent foundations, model organisms of misalignment, and human-computer interaction.

Yanda Chen is a member of technical staff at Anthropic and a PhD candidate in computer science at Georgetown University advised by Kevin Knight. His homepage says he previously worked at Allen Institute for AI and focuses on AI safety, natural language processing, and deep learning.

Dan Hendrycks is the executive director of the Center for AI Safety and an advisor to xAI and Scale AI. His public homepage also says he received his PhD in AI from UC Berkeley and highlights contributions including GELU, robustness benchmarks, and MMLU.

Dylan Hadfield-Menell is a research scientist at Anthropic. His homepage says he studies how to understand and align increasingly capable AI systems, with additional interests in game theory and inverse reinforcement learning. MIT CSAIL's profile says he completed a PhD in computer science at the University of California, Berkeley advised by Stuart Russell.

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

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 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.

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.

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.

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

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 cofounder of Oulipo Labs, working on language model safety, evaluations, and scientific forecasting.

Co-founder and CEO of Anthropic.

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

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

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

Assistant Professor of Philosophy at The University of Hong Kong and Research Fellow at Anthropic, working in ethics, epistemology, and social and political philosophy.

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.

Research scientist at Anthropic focused on safety and robustness for language models and reinforcement learning.

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.

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.

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.

Research scientist at Anthropic and assistant professor of computer science at Northeastern University working on interpretability and model understanding.

Anthropic report author listed on RLHF, Constitutional AI, Collective Constitutional AI, and Many-shot Jailbreaking reports, with report-backed work on alignment and adversarial evaluation.

Josh Batson is a research scientist at Anthropic. Public descriptions of his work emphasize understanding how and why AI systems work, especially interpretability.

Chief Science Officer and Co-Founder of Anthropic, with public bios emphasizing scaling laws and large language models.

Member of Anthropic's Societal Impacts team, where she studies the real-world impacts of AI systems.

Research scientist at Anthropic working on machine learning and AI safety.

PhD student at the University of Oxford working on AI safety, including scalable oversight and interpretability.

Researcher at Anthropic with interests in machine learning, AI alignment, and economics.

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.

Zachary Witten is a member of technical staff at Anthropic.

Member of technical staff at Anthropic whose work focuses on language models, model understanding, and alignment.

Buck Shlegeris is a Member of Technical Staff at Anthropic whose public homepage focuses on AI safety, model evaluations, and alignment.

Will McCrostie is listed as an author of the Anthropic technical report Tracing the thoughts of a large language model.

Researcher at Anthropic and coauthor of the Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Researcher at Anthropic and coauthor of the Constitutional Classifiers report.

Researcher at Anthropic and coauthor of the Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

President of METR and former team member at Anthropic whose work focuses on evaluating and forecasting frontier AI capabilities.

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

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

Avital Oliver is listed as an author of the Anthropic technical report Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Benjamin Lermen is listed as an author of the Anthropic technical report Auditing language models for hidden objectives.

Chenyan Zhang is listed as an author of the Anthropic technical report Auditing language models for hidden objectives.

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

David McDougall is listed as an author of the Anthropic technical report Many-shot Jailbreaking.

Jordan Taylor is listed as an author of the Anthropic technical report Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Jules Christmann is listed as an author of the Anthropic technical report Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Mantas Mazeika is listed as an author of the Anthropic technical report Many-shot Jailbreaking.

Saffron Huang is listed as an author of the Anthropic technical report Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Shibani Santurkar is listed as an author of the Anthropic technical report Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

Tomás Riofrío is listed as an author of the Anthropic technical report Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming.

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

Member of technical staff at Anthropic focused on safe and reliable AI.

Assistant professor of marketing at Stanford Graduate School of Business whose research uses AI systems to study human decision-making and related machine learning questions.

Co-founder and president of Anthropic and writer of the Cold Takes blog.

Computer scientist at Anthropic focused on making advanced AI systems safe and beneficial.

Member of Technical Staff at Anthropic and researcher in neural circuits and mechanistic interpretability, building tools for understanding AI systems.

Assistant Professor of Computer Science at the University of Oxford whose research spans generalization, reasoning, and large language model agents.

Researcher focused on AI alignment, reasoning under uncertainty, and the long-term safety of advanced AI systems.

Research scientist at Anthropic focused on AI alignment, language model behavior, and scalable oversight.

Member of Technical Staff at Anthropic working on AI control, hidden objectives, alignment, and evaluations, with a background in language models, efficient training, and scientific machine learning.

Member of technical staff at Anthropic working on deep learning, mechanistic interpretability, and AI safety.

Amanda Askell

Deep Ganguli

Tom Henighan

Aengus Lynch

Evan Hubinger

Jacob Hilton

Sandipan Kundu

William Saunders

Cem Anil

Saurav Kadavath

Ryan Greenblatt

Liane Lovitt

Samuel Marks

Yanda Chen

Dan Hendrycks

Dylan Hadfield-Menell

Kamal Ndousse

Jack Clark

Newton Cheng

Jan Leike

Samuel R. Bowman

David Duvenaud

Ethan Perez

Nicholas Schiefer

Dario Amodei

Shauna Kravec

Nova DasSarma

Tom Conerly

Simon Goldstein

Jesse Mu

Linda Petrini

Roger Grosse

Jared D. Kaplan

Yuntao Bai

David Bau

Anna Chen

Josh Batson

Jared Kaplan

Carina Kauf

Sören Mindermann

Henry Sleight

Jack Chen

Kshitij Sachan

Michael Sellitto

Mrinank Sharma

Zachary Witten

Alex Tamkin

Buck Shlegeris

Will McCrostie

Cameron Raymond

Maxwell Tegmark

Valentin Meksyn

Zara Ahmed

Beth Barnes

Carson Denison

Monte MacDiarmid

Avital Oliver

Benjamin Lermen

Chenyan Zhang

Daniel M. Ziegler

David McDougall

Jordan Taylor

Jules Christmann

Mantas Mazeika

Saffron Huang

Shibani Santurkar

Tomás Riofrío

Adam Jermyn

Alexey Nazarov

Esin Durmus

Holden Karnofsky

Jan Brauner

Johannes Treutlein

Owain Evans

Paul Christiano

Rylan Schaeffer

Scott Emmons

Wes Gurnee