# AI Has Begun Producing Novel Mathematics Beyond Its Training Data by Emad Mostaque > As of the last month and a half, the models have gone beyond their training data definitively. Before Navier–Stokes, OpenAI released ten proofs of famous math problems, and at least two of them — [one] conjecture being the main one — [are] definitively novel, elegant, and never seen before, the way they did that proof. [AI solved] Navier–Stokes in a bit of a more mechanical way, but again in an incredibly novel way. [The Navier–Stokes proof] is quite difficult to read because it’s like AI slop Claudism when you’re reading it. It took three days for me to even read the paper, and I haven’t gone through the formal stuff. I’ve talked to enough of my mathematician buddies to say: yeah, a human could have done [the AI-generated Navier–Stokes proof] given twenty or thirty years. > **— Adapted from Emad Mostaque**, *Impact Theory with Tom Bilyeu, 2026* ## Sources and Context - **Recording and precise locator:** [5:43](https://www.youtube.com/watch?v=WsdcF7EEvhM&t=343s) — Impact Theory with Tom Bilyeu, 2026. - **Exact reconciled source object:** “As of the last month and a half, the models have gone beyond their training data definitively. Before Navier–Stokes, OpenAI released ten proofs of famous math problems, and at least two of them — [one] conjecture being the main one — [are] definitively novel, elegant, and never seen before, the way they did that proof. [AI solved] Navier–Stokes in a bit of a more mechanical way, but again in an incredibly novel way. [The Navier–Stokes proof] is quite difficult to read because it’s like AI slop Claudism when you’re reading it. It took three days for me to even read the paper, and I haven’t gone through the formal stuff. I’ve talked to enough of my mathematician buddies to say: yeah, a human could have done [the AI-generated Navier–Stokes proof] given twenty or thirty years.” - **Editorial status:** The poster is transparently labeled as an adaptation because the reconciled dossier required synthesis beyond bracket-only repair. - **Reconciled source dossier:** [[research/AI Is Now So Good They Will Pay You to Quit|AI Is Now So Good They Will Pay You to Quit]] preserves transcript order, speaker attribution, editorial conventions, and the surrounding argument. - **Weekly promotion dossier:** [[research/Weekly Reminder Hunt - 2026-10-02|Weekly Reminder Hunt — October 2, 2026]] records why this object was selected for permanent promotion. ## Related Articles and Collections - **Collection:** [[collections/Simple Reminders|Simple Reminders]] - **Collection:** [[collections/Machine Succession|Machine Succession]] - **Article:** [[articles/The Gigawatt Is the New Measure of War|The Gigawatt Is the New Measure of War]] ## Related Topics - [[wiki/Emad Mostaque|Emad Mostaque]] - [[wiki/Compute Race|Compute Race]] - [[wiki/Biological Moonshots|Biological Moonshots]] - [[wiki/Accelerated Science|Accelerated Science]] ## Share on Social Media ``` As of the last month and a half, the models have gone beyond their training data definitively. Before Navier–Stokes, OpenAI released ten proofs of famous math problems, and at least two of them — [one] conjecture being the main one — [are] definitively novel, elegant, and never seen before, the way they did that proof. [AI solved] Navier–Stokes in a bit of a more mechanical way, but again in an incredibly novel way. [The Navier–Stokes proof] is quite difficult to read because it’s like AI slop Claudism when you’re reading it. It took three days for me to even read the paper, and I haven’t gone through the formal stuff. I’ve talked to enough of my mathematician buddies to say: yeah, a human could have done [the AI-generated Navier–Stokes proof] given twenty or thirty years. — Adapted from Emad Mostaque, Impact Theory with Tom Bilyeu, 2026 https://bryantmcgill.com/simple-reminders-mostaque-ai-has-begun-producing-novel-mathematics-beyond ```