person1979–

Eliezer Yudkowsky

Disciplines:

Summary

American AI safety researcher — co-founder of the Machine Intelligence Research Institute (MIRI) — creator of LessWrong — author of Harry Potter and the Methods of Rationality and the Sequences — who has been arguing since 2000 that AI poses an existential risk and has produced the foundational framework for AI safety thinking.

Core Contribution

Creates the AI safety problem formulation — Yudkowsky's articulation of the alignment problem (that an AI optimizing for the wrong objective function will pursue it in ways catastrophic for humans), his concepts of instrumental convergence, and his insistence that the problem deserves systematic research defined the intellectual program of AI safety.

Key Spikes & Highlights

Has never attended a day of college and is the founder of the field of AI safety research — his self-education (beginning at 15, having conversations with Carl Sagan and others by email) is the most consequential non-institutional intellectual training in Silicon Valley history
Stopped publishing AI safety technical research around 2015 and publicly stated he thought the problem was likely unsolvable given current trajectories — his move from 'this is solvable' to 'I don't think we're going to make it' is the most dramatic public research pessimism statement in AI history; he has compared the situation to discovering you are on a plane with no pilots

Notable Works & Achievements

The Sequences

Harry Potter and the Methods of Rationality

Rationality: A-Z

Inadequate Equilibria

Historical Influences

Self-education Vernor Vinge Douglas Hofstadter

Intellectual Lineage

Seattle Self-educated MIRI LessWrong

Contemporaries & Peers

Nick Bostrom Scott Alexander Robin Hanson

Minds Influenced

AI safety field EA movement

Intellectual Posterity

AI alignment LessWrong MIRI Sequences Instrumental convergence Harry Potter and the Methods of Rationality

Discussing