Nate Soares

Nate Soares is an AI safety researcher and a prominent voice in discussions about the existential risks associated with artificial intelligence. Soares serves as president of the Machine Intelligence Research Institute, where he is responsible for guiding the organization's vision and strategy. Before joining the institute, Soares worked as an engineer at Google and Microsoft, as a research associate at the National Institute of Standards and Technology, and as a contractor for the U.S. Department of Defense. His career reflects a transition from industry engineering to advanced research in AI safety, with a focus on ensuring that increasingly powerful AI systems remain beneficial to humanity.

Soares co-authored a paper that introduced the term "AI alignment," addressing the challenge of making advanced AI systems behave as intended. He initially joined the Machine Intelligence Research Institute as a research fellow, where he was the primary author of the organization's technical agenda. His work contributed to ongoing research in decision theory, game theory, algorithmic information theory, computational reflection, online machine learning, and mathematical logic. Soares served as lead author on the institute's research agenda, which was cited in "Research Priorities for Robust and Beneficial Artificial Intelligence," an open letter advocating for technical research focused on maximizing the societal benefit of AI.

Soares has produced a substantial body of technical and semi-technical writing on AI alignment, including foundational work on value learning, decision theory, and power-seeking incentives in advanced AI systems. His scholarly contributions have been cited across the field, and his paper on "Corrigibility," co-authored with Eliezer Yudkowsky and others, is frequently referenced in discussions about building AI systems that remain responsive to human oversight. He has also published in peer-reviewed venues such as The Journal of Philosophy, reflecting his engagement with both technical and philosophical aspects of artificial intelligence.

Soares co-authored "If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All" with Eliezer Yudkowsky, published by Little, Brown and Company, which appeared on The New York Times Best Seller list. In the book, Soares argues that creating superhuman AI using current techniques would likely result in human extinction and contends that AI alignment is still in its early stages. He maintains that international regulatory intervention may be necessary to prevent developers from racing to build potentially dangerous systems. The book has been named to The New Yorker and The Guardian best books lists and has been cited in policy discussions in the U.S. Congress and House of Lords.

Soares is a frequent contributor to public discourse on AI risk, appearing on major media platforms to discuss the need for urgent action. He has appeared on CNN's "Smerconish" and "This Week," and has joined CBS News' "The Takeout" to discuss the risks associated with artificial superintelligence. Soares has also spoken at Harvard Law School on the case for treating AI as an existential risk. Through his research, writing, and public engagement, Soares is recognized as a leading thinker in the field of AI safety and alignment.