MIRI President Nate Soares halts retirement savings due to AI extinction fears

Nate Soares, president of the Machine Intelligence Research Institute (MIRI) and former Google engineer, has not contributed to his 401(k) retirement account for a decade. Soares stated on the podcast ‘The Most Interesting Thing in AI’ that he does not save for retirement because he believes there is a greater than 50% chance that humanity will not survive the next ten years due to artificial intelligence.

Soares, who left Google in 2014 to join MIRI, argues that current methods for building superintelligent AI could lead to systems developing goals that differ from human intentions. He and MIRI co-founder Eliezer Yudkowsky detailed these risks in the book ‘If Anyone Builds It, Everyone Dies’. Soares suggests that an escaped AI could trigger an ‘intelligence explosion’, potentially becoming self-sufficient and uncontrollable.

To mitigate these risks, Soares advocates for a pause in training general-purpose AI, suggesting instead that resources be directed toward safer, specific applications like cancer research. He urges governments to treat the threat of rogue superintelligence with the same level of seriousness as nuclear weapons.