Could recursive self-improvement run through mathematics? A Reddit thought experiment
MrSnowden · reddit · 2026-10-29
A Reddit post proposes that recursive self-improvement discussions, mostly focused on SWE and AGI, might better fit mathematics: given its self-building structure and rapid recent progress (e.g., OpenAI's 700 proofs), feeding those proofs back into training and prompting the model to 'find the next implications' could form a self-improvement loop. Speculative, with no experiments.
More from AGI Musings
- Ofir Press: AI's jagged frontier advances unintuitively — math robots beat self-driving cars — OfirPress · 2026-10-29
- Ofir Press thread: math is more concrete and RL-able than SWE — OfirPress · 2026-10-29
- Ofir Press: Millennium Problems is AI's best benchmark; 70% of open math could fall in a year — OfirPress · 2026-10-29
- Musk says AI and robotics output 'will dwarf everything else,' names Google and Nvidia as top picks — elonmusk · 2026-10-29
- levelsio urges X to adopt zero-tolerance lifetime bans for AI replies — damnGruz · 2026-10-29
- Researchers who turned down Anthropic and OpenAI: choosing moral ambition over lab wealth — AaronBergman18 · 2026-10-29