An OpenAI model has disproved a central conjecture in discrete geometry

OpenAI의 내부 AI 모델이 80년간 미해결이던 조합기하학의 핵심 문제인 평면 내 단위 거리 쌍 최대 개수 문제를 해결하며 기존의 정설을 반박했다. 이 증명은 수학자들에 의해 검증되었고, AI가 독창적이고 복잡한 수학적 아이디어를 자율적으로 도출해낸 첫 사례로 평가받는다. 특히 대수적 수 이론의 고급 개념을 활용해 기존의 정수론적 접근을 확장한 점이 주목된다. 이 성과는 AI와 수학 연구의 새로운 협업 가능성을 보여주며, AI가 단순 보조를 넘어 창의적 연구자로 자리매김했음을 시사한다.

https://openai.com/index/model-disproves-discrete-geometry-conjecture/

#openai #aimathematics #discretegeometry #algebraicnumbertheory #llm

An OpenAI model has disproved a central conjecture in discrete geometry

An OpenAI model solved the 80-year-old unit distance problem, disproving a major conjecture in discrete geometry and marking a milestone in AI-driven mathematics.

OpenAI
An OpenAI model has disproved a central conjecture in discrete geometry

An OpenAI model solved the 80-year-old unit distance problem, disproving a major conjecture in discrete geometry and marking a milestone in AI-driven mathematics.

OpenAI

Inspired by the No-three-in-line-Problem https://en.wikipedia.org/wiki/No-three-in-line_problem i studied a much simpler case and came up with a mini-conjecture:

Definition: A position on a Grid is called "blocked" if there would be three in Line, if a point is placed there.

Conjecture: You can't place n points on a n x n Grid without blocking a point or putting 3 in a line for n > 4.

Any hints on proving or disproving it? I am pretty sure it is true.

#math #NoThreeInLineProblem #DiscreteGeometry

(edited after comments)

No-three-in-line problem - Wikipedia