Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

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

Co-founder and head of policy at Anthropic. He previously served as policy director at OpenAI, worked as a technology journalist, and writes the Import AI newsletter.

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

Staff software engineer at Anthropic building systems for AI safety, reliability, and alignment.

Anthropic researcher working on machine learning and AI-assisted science; previously built tools for learning from text, images, and tabular data.

Research scientist known for mechanistic interpretability and deep learning visualization, previously at Google Brain and OpenAI.

Alignment researcher at OpenAI working on making AI understandable to and aligned with human values.

Anthropic co-founder and Chief Science Officer. Formerly a physicist at Johns Hopkins, he helped develop scaling laws for neural language models and works on the science and safety of large AI systems.

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.

Independent researcher working on the theoretical foundations of AI, especially inductive biases, scaling laws, and approximate Bayesian updating. His public homepage notes prior research roles at Anthropic and OpenAI.

Member of Anthropic's Interpretability team, where he works on understanding how large language models work.

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

Catherine Olsson is an AI alignment researcher and writer whose public website and Anthropic author page describe work on AI safety, interpretability, and building helpful, harmless assistants.

Co-founder and head of alignment science at Anthropic.

CEO and co-founder of Anthropic. Before Anthropic, he served as vice president of research at OpenAI.

Research scientist at Anthropic interested in understanding neural networks and applying that understanding to alignment.

Researcher working on AI safety and adversarial evaluation, including Anthropic many-shot jailbreaking research.

Research scientist at Anthropic interested in understanding and steering AI systems.

Software engineer at Anthropic, previously at Google, with public writing on language models, agents, and reinforcement learning.

Researcher interested in neural networks and their potential to achieve general intelligence. His public homepage notes prior roles as a cofounder at Anthropic, researcher at OpenAI, and member of the startup team at Stripe.

Researcher at Anthropic working on the alignment and evaluation of advanced AI systems.

Research scientist at Anthropic working on model behavior and interpretability.

Software engineer at Anthropic working on infrastructure, tooling, model behavior, and multimodal systems.

Member of Technical Staff at Anthropic whose work focuses on understanding, evaluating, and improving large language models, with emphasis on reasoning, safety, and generalization.