Jathan Sadowski recently recommended a paper by Alexander Campolo and Katia Schwerzmann, “From rules to examples: Machine learning's type of authority,” which considers the epistemological implications of the “transition from a rule-based computer programming paradigm to an example-based paradigm associated with machine learning” (also a key topic for Matteo Pasquinelli’s recent book
this is such a great breakdown and expounding upon of some of the really cool papers that have been coming out on the subject! I read both the Bender/Shah paper and the Campolo/Schwerzmann piece when they came out, and I feel I get the crux of both a little better having read your essay — thank you!
what this relates to for me is the idea of “AI as a floating signifier” — this gimmick that even if these generative AI technologies are not “real” models of the world TODAY, with enough data and power, they will be SOON!
❤️🔥
this is such a great breakdown and expounding upon of some of the really cool papers that have been coming out on the subject! I read both the Bender/Shah paper and the Campolo/Schwerzmann piece when they came out, and I feel I get the crux of both a little better having read your essay — thank you!
also in this realm is Lucy Suchman’s recent work (although maybe you’ve come across it already): https://journals.sagepub.com/doi/full/10.1177/20539517231206794
what this relates to for me is the idea of “AI as a floating signifier” — this gimmick that even if these generative AI technologies are not “real” models of the world TODAY, with enough data and power, they will be SOON!