Answers are grounded in this paper and cite their sources.
This research draft presents a proposed proof that, for every zeta > 0 and all sufficiently large n, the n by n integer grid contains a subset of at least (5/2 - zeta)n points with no four distinct points on a common circle or a common line. Combined with the known upper bound floor((5n - 3)/2), the claimed lower bound would determine the asymptotic coefficient as 5/2. The construction uses two correlated old layers and a third layer selected by a hypergraph matching. Axis correlation kernels, diagonal presence-state allocations, general counting statements, and a rounding and deletion argument connect the continuous capacity calculations to finite point sets. The release contains a single 34-page manuscript, editable Markdown and standalone LaTeX sources, Python checking programs, fixed integer arrays, an exact rational analytic certificate, execution instructions, and a SHA-256 manifest. The finite computations certify their specified analytic bounds and finite identities; they do not by themselves verify the general probabilistic counting argument. This is a proposed proof, and no formal proof-assistant verification is claimed. AI assistance was substantial: OpenAI Codex was used for mathematical ideas, derivations, proof drafting, numerical searches, writing and executing checking code, textual checks, editing, and preparation of the publication files. Revisions also incorporate checks and calculations from separate GPT and Claude sessions. The package contains no named author.
No takes yet. Share an insight, caveat, or question.
Anonymous (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: