Nate Soares

In an interview with Max Raskin, Soares recalled growing up in rural Vermont and reading science fiction including Ender’s Game and works by Vernor Vinge.⁠[1]

He earned a BS in computer science and economics from George Washington University.⁠[2]

Before joining MIRI, he worked as a software engineer at Google and Microsoft, a research associate at the National Institute of Standards and Technology (NIST), and a contractor for the US Department of Defense.⁠[3]

While working at GoogleGoogleGoogle is an American technology company founded by Larry Page and Sergey Brin and incorporated in 1998. It became an Alphabet subsidiary in 2015. Its businesses include internet search, advertising, software platforms, cloud computing, and artificial intelligence.Open the full entry, Soares attended his first MIRI workshop in December 2013 and joined its research team in 2014. His 2015 introduction described his work on the research agenda and technical papers.⁠[4]

With Benja Fallenstein, Eliezer Yudkowsky and Stuart Armstrong, he co-authored Corrigibility, examining how AI systems could accept human correction or shutdown without trying to prevent or cause shutdown. The paper was accepted by the AI and Ethics workshop at AAAI 2015.⁠[5]

In June 2015, he succeeded Luke Muehlhauser as MIRI’s executive director. Before taking the role, he identified growing the research team as a top priority.⁠[4]

He and Benja Fallenstein co-authored Toward Idealized Decision Theory, discussing limitations of conventional decision theories concerning policy selection and logical counterfactuals and identifying problems for further formalization.⁠[6]

He co-authored Logical Induction, which studies how a computationally limited reasoner can assign probabilities to mathematical statements before they are proved or disproved and refine those judgments as reasoning progresses.⁠[7]

With Eliezer Yudkowsky, he proposed functional decision theory, treating decisions as outputs of mathematical functions and comparing the approach with causal and evidential decision theories. The work concerns formal decision theory.⁠[8]

On October 10, 2023, MIRI announced a leadership restructuring: Malo Bourgon became CEO, while Soares moved from executive director to president and remained a board member. Soares focused on strategy and research as MIRI increased its emphasis on public communication.⁠[9]

On September 16, 2025, his book with Eliezer Yudkowsky, If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, was published. The authors argue that superhuman AI poses an irreversible existential danger and that its development should be prevented; this is their position.⁠[10]

In The Atlantic, Adam Becker challenged the book’s assumptions about superintelligence and argued that doomsday narratives distract from present AI harms.⁠[11]

He and Greg Kestin released a preprint comparing AI-risk judgments before and after a public event at Harvard UniversityHarvard UniversityHarvard University is a private research university in Massachusetts, established by the Massachusetts Bay Colony legislature in 1636 and named in 1639 for early benefactor John Harvard. Its principal campuses are in Cambridge and the Allston and Longwood neighborhoods of Boston. It includes Harvard College, graduate and professional schools, and the Radcliffe Institute for Advanced Study. Two governing boards share oversight. Alan M. Garber was president as of October 2026, and the endowment stood at $56.9 billion at the end of fiscal 2025.Open the full entry, using 89 matched respondents. It reported increased perceived risk afterward, measuring changes in self-selected attendees’ views rather than actual extinction probability or public opinion overall.⁠[12]

In an October 24 interview post, Nicholas Thompson said Soares considered AI-driven human extinction by 2036 more likely than not and was not currently saving for retirement. These were Soares’s subjective risk assessment and personal choice, not an established prediction of future events.⁠[13]

On November 1, Soares responded to Mustafa SuleymanMustafa SuleymanMustafa Suleyman is a British AI entrepreneur and CEO of Microsoft AI who co-founded DeepMind and Inflection AI.Open the full entry’s discussion of AI consciousness. He agreed that humanity was unprepared for conscious machines but expressed less confidence that none had already been built.⁠[14]

He writes the personal blog Minding Our Way, including the Replacing Guilt series, and states that the blog does not represent MIRI’s views.⁠[15]