I am Co-Founder & Policy Analyst at Secure AI Project. I am also a master's student in Security Studies at Georgetown University. Previously, I worked at the Center for AI Safety Action Fund on California Senate Bill 1047.
Before that, I was a Junior Fellow at CSET. I did my undergrad in computer science at Yale. During a gap semester, I was the first employee at the Center for AI Safety.
This is my personal website, and it doesn't necessarily represent the opinions of my past or present employers. My past writings may no longer reflect my present views.
This website was last updated February 6 2025
It used to take me a long time to design a website like this. I did it in 30 minutes with an LLM.
Links
- Twitter/X
- Google Scholar
- Substack
- Beetss (I created the beet emoji 🫜)
- Randomness (test how random you are)
Posts
- Evaluating Large Language Models (CSET blog, July 2024, with Helen Toner)
- Emergent Abilities in Large Language Models: An Explainer (CSET blog, April 2024)
- Multimodality, Tool Use, and Autonomous Agents: Large Language Models Explained, Part 3 (CSET blog, March 2024, with Helen Toner)
- How Developers Steer Language Model Outputs: Large Language Models Explained, Part 2 (CSET blog, March 2024, with Helen Toner)
- Keeping Up with the Frontier: Why Congress Should Codify Reporting Requirements For Advanced AI Systems (CSET blog, February 2024)
- Devising ML Metrics (CAIS blog, March 2024, with Dan Hendrycks, originally written 2022)
Papers
- Helen Toner, John Bansemer, Kyle Crichton, Matthew Burtell, Thomas Woodside, Anat Lior, Andrew Lohn, Ashwin Acharya, Beba Cibralic, Chris Painter, Cullen O’Keefe, Iason Gabriel, Kathleen Fisher, Ketan Ramakrishnan, Krystal Jackson, Noam Kolt, Rebecca Crootof, and Samrat Chatterjee, Through the Chat Window and Into the Real World: Preparing for AI Agents (2024, CSET Workshop Paper)
- Noam Kolt, Markus Anderljung, Joslyn Barnhart, Asher Brass, Kevin Esvelt, Gillian K. Hadfield, Lennart Heim, Mikel Rodriguez, Jonas B. Sandbrink, and Thomas Woodside, Responsible Reporting for Frontier AI Development (2024, Conference on Artificial Intelligence, Ethics, and Society)
- Dan Hendrycks, Mantas Mazeika, and Thomas Woodside, An Overview of Catastrophic AI Risks (2023, arxiv preprint)
- Alexander Pan, Chan Jun Shern, Andy Zou, Nathaniel Li, Steven Basart, Thomas Woodside, Jonathan Ng, Hanlin Zhang, Scott Emmons, and Dan Hendrycks, Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark (2023, International Conference on Machine Learning)
- Steven H Wang, Antoine Scardigli, Leonard Tang, Wei Chen, Dimitry Levkin, Anya Chen, Spencer Ball, Thomas Woodside, Oliver Zhang, and Dan Hendrycks, MAUD: An Expert-Annotated Legal NLP Dataset for Merger Agreement Understanding (2023, Empirical Methods in Natural Language Processing)
- Matthew Burtell and Thomas Woodside, Artificial Influence: An Analysis of AI-Driven Persuasion (2023, arxiv preprint)
Class papers
- A Mathematical Model Of Dissent (2023, Philosophy of Dissent)
- Investigating Trojan Attacks In Large Language Models (2022, Senior Thesis)
- A Simulator Philosophy Of Mind (2022, Philosophy of Mind)
- Is Non-Self a Problem for Buddhist Ethics? (2022, Buddhist Thought)
- Artificial Influence: An Analysis Of AI-Driven Persuasion (2022, with Matt Burtell, Computer Science and Global Affairs)
- A Creepy Feeling: Nixon's Decision to Disavow Biological Weapons (2021, The American Century)
- Home Prices And Rents During COVID-19 (2021, Economics of Cities)
- Controlling A Video Game With an EEG (2019, Applied Science Research)
- Conscious Of Receding Power: The Partisan Motivations Behind The Single-Member Districting Mandate of 1842 (2018, AP US History)