Daniel Fried

Daniel Fried is an assistant professor in the Language Technologies Institute at Carnegie Mellon University. His research focuses on NLP, grounding and interaction, and the strategic use of language, with a particular focus on language interfaces such as LLM agents and code generation. Previously, he was a postdoc at Meta AI and the University of Washington and completed a PhD at UC Berkeley. His research has been supported by an NSF CAREER Award, a Microsoft Faculty Fellowship, and an Okawa Research Award.

AI2050 Project

AI agents increasingly work in teams: for example collaborating on software, negotiating on behalf of users, or sharing infrastructure. In teams of heterogenous agents, it’s crucial to verify that partners follow through on promises, communicate honestly, and remain understandable to human overseers. Fried’s project develops methods for trustworthy multi-agent communication: distributed verification where agents with different expertise collaborate to check commitments, training that produces faithful communication grounded in explicit plans, oversight agents that make the verification layer interpretable to humans, and adaptive models that let agents rapidly assess new partners. This lays the groundwork for reliable, human-compatible multi-agent AI.