AI agents sometimes produce unreadable but correct math
burny_tech · x · 2026-11-18
burnytech observed that AI agents sometimes produce math that is correct but unreadable to humans — the model gets the right answer while writing derivations in incomprehensible notation, dubbed "mathelese." An intriguing window into how agents reason internally on long math chains.
More from Models
- Code model eval: Rust hits 95-100% accuracy as 13 languages see big gains — shuding · 2026-11-18
- Asked Opus to make a video explaining consciousness and existence — "no wiser but cool" — flowersslop · 2026-11-18
- theo retests GPT-6.2 Astra: better, but still 'directionally very bad' at debugging — airesearch12 · 2026-11-18
- Claude correctly folds ubiquitin from a plain prompt, no specialized tools needed — iScienceLuvr · 2026-11-18
- GPT-6.2 Astra Tops Bug-Hunt Coding Benchmark at 62.7/105, One-Fifth Sonnet's Cost — PawelHuryn · 2026-11-18
- OpenAI's enterprise Dots reportedly to cost $50/$200/$500 per tier plus usage — and businesses are turning them off — jdjohnson · 2026-11-18