about me
I am a research scientist at Anthropic, where I started and lead the societalimpacts team. Prior to that, I was the founding research director at the Stanford Institute for Human Centered AI (HAI). I did my PhD in Computational Neuroscience at NYU and obtained my BS in Electrical Engineering and Computer Science (EECS) from Berkeley. For fun, I surf, play bass with my band, and read widely across the humanities and social sciences. Here's some of my and some
research
Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions
COLMlast author
Apr 2025Many-shot Jailbreaking
NeurIPS
Dec 2024The Capacity for Moral Self-Correction in Large Language Models
arXivfirst author
Feb 2023Red Teaming Language Models to Reduce Harms
arXivfirst author
Aug 2022Constitutional AI: Harmlessness from AI Feedback
arXivmiddle author - evals
Dec 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
arXivmiddle author - evals
Apr 2022
Evaluating and Mitigating Discrimination in Language Model Decisions
NeurIPS Algorithmic Fairness Workshoplast author
Dec 2024Evaluating Feature Steering: A Case Study in Mitigating Social Biases
Anthropic Research Bloglast author
Oct 2024Towards Measuring the Representation of Subjective Global Opinions in Language Models
COLMlast author
Oct 2024Measuring the Persuasiveness of Language Models
Anthropic Research Bloglast author
Apr 2024Challenges in Evaluating AI Systems
Anthropic Research Blogfirst author
Oct 2023Discovering Language Model Behaviors with Model-Written Evaluations
ACLmiddle author - evals
July 2023Beyond the Imitation Game: Quantifying and Extrapolating the Capabilities of Language Models
TMLRmiddle author - evals
May 2022
Introducing Anthropic Interviewer: What 1,250 professionals told us about working with AI
Anthropic Research Bloglast author
Dec 2025How AI Is Transforming Work at Anthropic
Anthropic Research Bloglast author
Dec 2025Which Economic Tasks are Performed with AI? Evidence from Millions of Claude Conversations
arXivlast author
Feb 2025Anthropic Economic Index: Understanding AI's Effects on the Economy Over Time
Anthropic Economic Indexlast author
Feb 2025
The Impact of Advanced AI Systems on Democracy
Nature Human Behaviourmiddle author
Oct 2025Collective Constitutional AI: Aligning a Language Model with Public Input
FAccTlast author
Jun 2024Testing and Mitigating Elections-related Risks from AI
Anthropic Research Blog
Jun 2024Opportunities and Risks of LLMs for Scalable Deliberation with Polis
arXivlast author
Jun 2023
How People Use Claude for Support, Advice, and Companionship
Anthropic Research Bloglast author
Jun 2025Predictability and Surprise in Large Generative Models
FAccTfirst author
Jun 2022The AI Index 2021 Annual Report
arXiv
Mar 2021Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models
arXivlast author
Feb 2021
Neural and Perceptual Signatures of Efficient Sensory Coding
arXivfirst author
Feb 2016Efficient Sensory Encoding and Bayesian Inference with Heterogeneous Neural Populations
Neural Computationfirst author
Oct 2014Implicit Encoding of Prior Probabilities in Optimal Neural Populations
NeurIPSfirst author
Dec 2010
media
It's their job to keep AI from destroying everything
The Verge
Dec 2025Societal Impacts Team
What if We Could All Control AI?
The New York Times
Oct 2023Collective Constitutional AI: Aligning a Language Model with Public Input
The 3 Most Important AI Innovations of 2023
Time Magazine
Dec 2023Collective Constitutional AI: Aligning a Language Model with Public Input
What if Dario Amodei is Right About AI?
The Ezra Klein Show
Apr 2024Measuring the Persuasiveness of Language Models
The Unpredictable Abilities Emerging From Large AI Models
Quanta Magazine
Mar 2023Predictability and Surprise in Large Generative Models
I Used Anthropic's Interviewer Tool to Share My AI Complaints
ZDNet
Dec 2025Introducing Anthropic Interviewer: What 1,250 professionals told us about working with AI
Anthropic Will Start Using AI to Interview Its Users About Their Experience with AI
The Verge
Dec 2025Introducing Anthropic Interviewer: What 1,250 professionals told us about working with AI
Anthropic Turns Inward to Show How AI Affects Its Own Workforce
Inc
Dec 2025How AI Is Transforming Work at Anthropic
AI is making the workplace lonelier
Axios
Dec 2025How AI Is Transforming Work at Anthropic
Anthropic just analyzed 700,000 Claude conversations and found its AI has a moral code of its own
VentureBeat
Apr 2025Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions
Exclusive: Anthropic's Index Tracks AI Economy
Axios
Feb 2025Which Economic Tasks are Performed with AI? Evidence from Millions of Claude Conversations
How Claude Uses AI to Identify New Threats
Platformer
Dec 2024Clio: Privacy-Preserving Insights into Real-World AI Use
Anthropic Leads Charge Against AI Bias and Discrimination With New Research
VentureBeat
Dec 2023Evaluating and Mitigating Discrimination in Language Model Decisions
When Hackers Descended to Test AI, They Found Flaws Aplenty
The New York Times
Aug 2023Red Teaming Language Models to Reduce Harms
Language Models Might be Able to Self-correct Biases—If You Ask Them
MIT Technology Review
Mar 2023The Capacity for Moral Self-Correction in Large Language Models
OpenAI & Stanford Researchers Call for Urgent Action to Address Harms of LLMs Like GPT-3
VentureBeat
Feb 2021Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models
Starfish Enterprise: Finding RNA Patterns in Single Cells
Nature
Aug 2019Starfish: Open Source Image Based Transcriptomics and Proteomics Tools