At ICM 2026, researchers discussed how LLMs are reshaping proof formalization
AlexKontorovich · x · 2026-07-26
- At ICM 2026, Alex Kontorovich and Scott Armstrong discussed autoformalization and the growing role of LLMs such as Claude and ChatGPT in proof formalization.
- The quoted segment says the shift is already underway: even without a Lean background, researchers are using LLMs to formalize results such as De Giorgi–Nash–Moser theory.
- The conversation emphasized a healthy diversity of views, but the direction is clear: Lean is becoming more central in mathematical formalization workflows.
More from AGI Musings
- Tyler Cowen says teenage AI maniacs will shape the future economy — nateliason · 2026-07-26
- Open-source Chinese AI models are forcing frontier labs to compete, not monopolize — bindureddy · 2026-07-26
- AI debate framed as a replay of Industrial Revolution-era reactions — joshua_saxe · 2026-07-26
- The model matters less than what you do with it, says AI watcher — StewartalsopIII · 2026-07-26
- Forced AI adoption is creating a market for opt-out workshops — HaktanSuren · 2026-07-26
- Anthropic’s global-workspace paper hints at a fixed semantic coordinate system in LLMs — tszzl · 2026-07-26