AI cracked the proof. Now what's a human?
On September 5, a fluid broke in the Navier-Stokes equations, a mathematical model describing fluid dynamics. An OpenAI system, more powerful than its latest GPT-6 Astra model, cracked the proof after 10,000 software agents worked for 88 hours. Two mathematicians, Tristan Buckmaster and Levent Alpoge, spent a year on the problem, following an idea from Spanish mathematicians Diego Cordoba and Luis Martinez-Zoroa.
However, an even more powerful machine found the proof first. A proof exists, but it is beyond human understanding. The machine's process is opaque, and the answer and understanding have been separated. The implications are profound: answers are becoming cheap, while questions are becoming expensive; we risk losing the ability to check proofs; disputes will shift from truth to names and contributions; and machines have made knowledge creation a new job of holding rather than making.
Written by urgent.news from The Economic Times's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
