Petty's projection conjecture in dimensions at least four
For a convex body the projection body has support function . The ratio is affine invariant. Petty (1971) conjectured that for it is minimized exactly by ellipsoids, i.e. with equality only for ellipsoids. Local versions were proved by Saroglou-Zvavitch and Ivaki, and in August 2026 Chen, Feng, Li, Xi and Xu proved the case . Does Petty's inequality, with its ellipsoid equality case, hold for every convex body in every dimension ?
- Result
- Proved(see note)
- Status
- Partial result
- AI contribution
- AI-discovered
- Method
- Argument
- Field
- Convex geometry; affine isoperimetric inequalities, projection bodies
- Posed by
- C. M. Petty, Isoperimetric problems, Proc. Conf. Convexity and Combinatorial Geometry, Univ. of Oklahoma (1971)
- Year posed
- 1971
- Years open
- 55y
- Solved
- 2026-09-24
- Model
- Unreleased internal OpenAI model
- Vendor
- OpenAI
- Collaborators
- —
- Verification
- Lean-checked, statement unaudited
- Publication
- Announced
- Collection
- OpenAI math release (October 2026), version adc7f12
- Significance
- 42 / 100
- Disclosed cost
- —
- Wikipedia
- No dedicated article
What was actually shown
Theorem 1: for every and every convex body , , with equality iff is an ellipsoid; no symmetry or regularity is assumed. The proof bounds a bilinear pairing via a strict spherical-harmonic estimate for norms and a fixed-point choice of affine position. It does not treat , which Chen, Feng, Li, Xi and Xu proved in August 2026; the paper notes that the two together give the full conjecture for . As corollaries (via Lutwak's known implication) it states the Lutwak-Petty lower-degree inequalities for , without equality cases, and a sharp Holmes-Thompson isoperimetric inequality.
What the AI did
The release README says all results were produced by an unreleased internal OpenAI model with a fixed procedure, on average about three hours of ChatGPT Pro thinking compute per result. This result is not among the README's exceptions (the Hodge conjecture for CM abelian varieties and the Re(s) > 11/12 zero-free region). The manuscript is authored 'OpenAI' and names no human author.
Verification
No independent mathematician has checked this yet. Checked here: Theorem 1 of the manuscript was read against Petty's conjecture as the paper states it with the 1971 reference; it claims the inequality and the ellipsoid-only equality case for every convex body in every dimension . The proof was not refereed. The Lean challenge ComparatorChallenges/PettyProjectionVolume.lean (listed in lean/formalization.yaml, declaration OAI.PettyProjection.petty_projection_volume) was read here: for and every compact convex with nonempty interior it asserts pettyConstant and equality iff is an affine image of the unit ball, with defined through orthogonal shadow volumes. This states the headline claim. Not rebuilt here. The three-dimensional case is not part of this work and rests on the human preprint of Chen et al.