Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming

Co-founder and CEO of Anthropic.

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.

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

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

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.

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.

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

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.

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.

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

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.

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

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.

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.

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.

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.

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

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.

Researcher at Anthropic and coauthor of the 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.

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.

Canonical link

Dario Amodei

Amanda Askell

Jack Clark

Jared Kaplan

Kamal Ndousse

Deep Ganguli

Tom Henighan

Ethan Perez

Nicholas Schiefer

Samuel Marks

Jan Leike

William Saunders

Alex Tamkin

Yanda Chen

Beth Barnes

Jacob Hilton

Dylan Hadfield-Menell

Aengus Lynch

Avital Oliver

Cameron Raymond

Saffron Huang

Zara Ahmed

Jules Christmann

Shibani Santurkar