Kakeya Conjecture
A conjecture in mathematics concerning the dimension of Kakeya sets, asserting that any set in n-dimensional space containing a unit line segment in every direction must have full Hausdorff dimension n. It is one of the central unsolved problems in harmonic analysis and geometric measure theory.
Timeline (last 90 days)
Two manuscripts numbered 074 in OpenAI's release claimed to prove a stronger version of the three-dimensional Kakeya maximal function conjecture and the four-dimensional Kakeya conjecture respectively, totaling 272 pages and authored solely under the name OpenAI.
The Kakeya conjecture was posed by Japanese mathematician Soichi Kakeya in 1917, asking what is the minimum area required on a plane to turn a needle of length 1 around a full rotation.
Besicovitch proved that the area of a Kakeya set can be arbitrarily small or even zero, while still accommodating a unit-length needle in every direction.
The planar case of the Kakeya conjecture was proved by Davies in 1971, the three-dimensional case was resolved by Hong Wang and Joshua Zahl, and the four-dimensional case and above had not been proved by anyone before.
OpenAI's result numbered 074 only proves the four-dimensional Hausdorff dimension version, does not resolve the four-dimensional maximal function conjecture or five dimensions and above, and has neither Lean formalization nor peer review.
All Facts (5)
The Kakeya conjecture was posed by Japanese mathematician Soichi Kakeya in 1917, asking what is the minimum area required on a plane to turn a needle of length 1 around a full rotation.
50%UnverifiedBesicovitch proved that the area of a Kakeya set can be arbitrarily small or even zero, while still accommodating a unit-length needle in every direction.
50%UnverifiedThe planar case of the Kakeya conjecture was proved by Davies in 1971, the three-dimensional case was resolved by Hong Wang and Joshua Zahl, and the four-dimensional case and above had not been proved by anyone before.
50%UnverifiedOpenAI's result numbered 074 only proves the four-dimensional Hausdorff dimension version, does not resolve the four-dimensional maximal function conjecture or five dimensions and above, and has neither Lean formalization nor peer review.
50%UnverifiedTwo manuscripts numbered 074 in OpenAI's release claimed to prove a stronger version of the three-dimensional Kakeya maximal function conjecture and the four-dimensional Kakeya conjecture respectively, totaling 272 pages and authored solely under the name OpenAI.
50%