Erdős Problem #846 — Let A⊂R2A\subset \mathbb{R}^2 be an infinite set for which there exists some ϵ>0\epsilon>0 such that in any subset of AA of size nn there are always at least ϵn\epsilon n with no three on a line.

About 34 years old · traced to

Let A⊂R2A\subset \mathbb{R}^2 be an infinite set for which there exists some ϵ>0\epsilon>0 such that in any subset of AA of size nn there are always at least ϵn\epsilon n with no three on a line. Is it true that AA is the union of a finite number of sets where no three are on a line?

References

Progress summary

Refreshed
Claimed solved

A concrete infinite set has disproved the proposed finite-decomposition principle, so the problem is now solved negatively.

Erdős, Nešetřil, and Rödl asked whether an infinite A⊂R2A\subset\mathbb{R}^2 in which every sufficiently large finite subset contains linearly many points with no three collinear must be a finite union of such sets.

February–May 2026 counterexample

The explicit construction maps each edge {xi,xj}\{x_i,x_j\} of K∞K_{\infty} to Pe=(xi+xj,xi2+xixj+xj2)P_e=(x_i+x_j,x_i^2+x_ix_j+x_j^2). Collinearity of three points is equivalent to a triangle among the corresponding edges. Every finite edge set has a bipartite subgraph with at least n/2n/2 edges, giving ϵ=1/2\epsilon=1/2, while infinite Ramsey theory forbids any finite partition into triangle-free color classes. Independent accounts attribute the construction to a DeepMind prover agent and an internal OpenAI model; arXiv papers provide the mathematical argument.

Current status (as of June 2026): The conjecture is disproved by an explicit construction with ϵ=1/2\epsilon=1/2; nothing remains open in the stated problem.

Sources

Solutions 0

No solutions have been posted yet.