Tom Cunningham works at METR on AI capabilities and economics. He previously worked at OpenAI on economic research, and before that as one of the most senior data scientists at OpenAI, Twitter, and Meta, consulting on content moderation, company strategy, anti-trust, advertiser behavior and auction design, network effects, and experimentation. He was a resident fellow at the Integrity Institute, working on content moderation and AI, and continues academic work on cognition and decision-making.
Since mid-2024, Cunningham has focused on the short- and long-run economic impact of AI, including co-authoring the 2025 working paper "How People Use ChatGPT" with researchers including David J. Deming and Aaron Chatterji. His content moderation work was cited extensively in the Congressional investigation into the January 6, 2020 riot, and he co-organized a workshop that produced an influential paper on ranking tradeoffs. At Meta, he pioneered matrix models for characterizing network effects and invented new methods for estimating experiment effects, published with Dominic Coey.
Cunningham holds an MRes/PhD in Economics from the London School of Economics and was a College Fellow at Harvard University. He has presented at MIT, Wharton, Yale, Chicago Booth, and the WWW and CODE conferences, and founded the Bay Area Tech-Economics Seminar Series.
