Mathematicians call AI's Yang-Mills proof 'slop': no mass gap, object may be trivial
LucaAmb · x · 2026-11-12
A harsh critique of an AI-produced Yang-Mills proof is circulating on X. The quoted post calls it "Yang-Mills slop": no mass gap is proven, and there's only "evidence" that the constructed object is Yang-Mills — an object that could also be a free theory, i.e. potentially trivial.
Luca Ambrosio, a leading mathematician in functional analysis and optimal transport, responds that this isn't really an argument about AI, but about the usefulness of normalization in physics — noting it's not obvious that normalization is useful at all.
A high-level technical pushback from the math/physics community on AI mathematical reasoning claims, worth tracking.
Related event: Experts Question OpenAI's 1000-Page Yang-Mills Proof, Mass Gap Unsolved(3 posts)→
More from AGI Musings
- Can open-source AI enable bioweapons? Safety debate flares on X again — PMinervini · 2026-11-13
- Microsoft outage sparks warning: a million AI agents can share one point of failure — tallmetommy · 2026-11-13
- Ravi Vakil on AI: nuanced and 'very cautiously optimistic' — littmath · 2026-11-13
- LMU Study: One Word Flip Flops How People Judge AI Consciousness — rkulidzan · 2026-11-13
- Eli Lilly CEO pushes back on Dario Amodei's 'ludicrous' medical AI claims — Noaidi · 2026-11-13
- Gary Marcus declares November the beginning of the end of the AI bubble — GaryMarcus · 2026-11-13