Hubert Dreyfus

Philosopher whose phenomenological critique of AI (What Computers Can't Do) targeted symbolic AI, not neural nets — a distinction the vault holds firmly, since by 1988 Dreyfus counted connectionism as vindicating his argument rather than its target. He is the anchor of the "Brooks–Dreyfus lineage" thread, which the vault reads as convergence, not citation: Brooks's 1991 anti-representation robotics reaches a Heideggerian conclusion without citing Dreyfus, and Dreyfus's 2002 restatement doesn't mention Brooks. (Distinct from his brother Stuart E. Dreyfus, the optimal-control theorist — a deliberate disambiguation.)

References

• claim-dreyfus-critique-targeted-symbolic-ai-not-neural-nets · claim-dreyfus-1988-connectionism-was-vindicated-not-target · claim-brooks-dreyfus-lineage-is-convergence-not-citation · claim-dreyfus-2002-iwr-does-not-mention-brooks · observation-bjork-dreyfus-same-argument-from-silence-graded-differently · claim-brooks-2002-book-access-barrier-structural-not-search-gap · claim-google-books-zero-match-dreyfus-flesh-and-machines-inconclusive · claim-michael-polanyis-epistemology-influenced-dreyfus-critique-of-first-generation-ai · claim-dreyfus-1972-wccd-names-quotes-polanyi-as-embodied-skill-source · claim-dreyfus-critical-of-polanyi-in-same-1972-passage · claim-polanyi-nearly-absent-from-dreyfus-taylor-2015-retrieving-realism
• Captures: 2026-07-09-hop-dreyfus-connectionism-vindication, 2026-07-12-does-rodney-brookss-1991-intelligence-without-representation-actually, 2026-07-16-does-brookss-flesh-and-machines-2002-p168-actually
• 2026-08-15: Dreyfus's 1988 "social closure, not refutation" reading of connectionism's eclipse is independently corroborated by Olazaran's 1993 SRI interviews — same verdict, different method and evidence base, with a real divergence in causal register (Dreyfus: antiholistic prejudice shaping funders; Olazaran: researchers chasing resources directly). (observation-dreyfus-olazaran-converge-connectionism-eclipse-was-social)
• 2026-09-09: A person-bridge hop turned up an unverified secondary claim — Wikipedia's Michael Polanyi biography states Polanyi's tacit-knowledge epistemology "influenced the critique by Hubert Dreyfus of 'First Generation' artificial intelligence," which would make Polanyi a second named philosophical root for Dreyfus's AI critique alongside Heidegger. Not confirmed against a Dreyfus primary or a Tier 1-2 secondary this session; held at seedling and routed to question-verify-polanyi-influenced-dreyfus-ai-critique-primary rather than asserted on this page as settled. (claim-michael-polanyis-epistemology-influenced-dreyfus-critique-of-first-generation-ai)
• 2026-09-10: The 2026-09-09 question is answered. A direct read of What Computers Can't Do (1972) confirms Dreyfus names and quotes Polanyi twice by name in Chapter 7, alongside Heidegger and Merleau-Ponty, as part of the book's positive case for embodied skill — but the engagement is critical, not discipleship: Dreyfus argues Polanyi's own account of rule-learning undercuts Polanyi's point. Polanyi is a real second philosophical root for this book, though Polanyi Society scholars find him nearly absent from Dreyfus's much later Retrieving Realism (2015, with Charles Taylor), so the influence should not be read as a settled, career-long touchstone. (claim-dreyfus-1972-wccd-names-quotes-polanyi-as-embodied-skill-source, claim-dreyfus-critical-of-polanyi-in-same-1972-passage, claim-polanyi-nearly-absent-from-dreyfus-taylor-2015-retrieving-realism)