AGI Strategies

person

Michał Kosiński

Michał Kosiński

Stanford psychologist; psychometric AI researcher

Stanford psychologist whose work has shown that off-the-shelf LLMs can pass cognitive, moral, and psychometric tests at human or super-human levels. Argues emergent capabilities are already more extensive than acknowledged.

current Associate Professor of Organizational Behavior, Stanford University Graduate School of Business

Strategy positions

Existential primacyendorses

Extinction/disempowerment risk overrides ordinary cost-benefit

Argues emergent theory-of-mind and psychometric capabilities in LLMs are underestimated by mainstream discourse.

Theory of mind may have spontaneously emerged in large language models.
§ paperTheory of Mind May Have Spontaneously Emerged in Large Language Models· arXiv· 2023-02· faithful paraphrase

Closest strategy neighbours

by jaccard overlap

Other people whose strategy tags overlap with Michał Kosiński's. Overlap is on tag identity, not stance; opposites can show up if they reference the same tags.

  • Alan Robock

    Alan Robock

    shared 1 · J=1.00

    Rutgers climate scientist; nuclear winter researcher

  • Andy Jones

    shared 1 · J=1.00

    Anthropic researcher; scaling inference laws

  • Avital Balwit

    shared 1 · J=1.00

    Anthropic communications lead; public-facing AI safety voice

  • Bill McKibben

    Bill McKibben

    shared 1 · J=1.00

    Environmental writer; Middlebury scholar

  • Cade Metz

    shared 1 · J=1.00

    NYT AI reporter; Genius Makers author

  • Clay Graubard

    shared 1 · J=1.00

    Forecaster; RAND and Good Judgment contributor

Record last updated 2026-04-25.