Erdős Problem #123
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: proved (not automatically the live result).
Subject areas: number theory.
Read this status correctly
The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.
The question in plain language
Let \(a,b,c>1\) be pairwise coprime integers. Can every sufficiently large positive integer be expressed as a sum of distinct elements of
such that no selected summand divides another? The property is called divisibility-free completeness, or \(d\) completeness. Ordinary additive completeness, in which some summands divide others, is weaker.
A small example
For the coprime bases \((2,3,5)\), the numbers \(2\), \(3\) and \(5\) are pairwise nondividing members of \(S\), and their sum equals \(10\). A sum containing \(2\) and \(6\) violates the divisibility restriction, since \(2\) divides \(6\). A single finite representation says nothing about representations of all sufficiently large integers.
The weaker hypothesis \(\gcd(a,b,c)=1\) is insufficient. The source gives \((6,10,15)\): modulo three, every allowed power product is zero unless it is a pure power of \(10\), which is one modulo three. A sum congruent to two therefore requires at least two pure powers of \(10\), and these divide one another.
What is known
PROVED in a published 2026 proof claim, with reported third-party Lean formalization. The maintained Erdős #123, edited 17 July 2026, attributes the affirmative answer to GPT-5.6 prompted by Colin Snyder. It links a mathematical proof summary. The current upstream Formal Conjectures entry credits a formalization by Boris Alexeev and identifies this third-party Lean proof snapshot.
The protected statement at revision 85f86371 was directly read. Its theorem quantifies over \(a,b,c>1\), pairwise coprimality, and a finite antichain of summands under divisibility. Its body contains a proof hole: existence of the protected statement is not itself proof execution. Earlier Erdős–Lewin work proved special cases, including \((3,5,7)\). An additional snugness problem for \((2,3,5)\) requires all selected summands to be close in size; the general \(d\) completeness theorem does not settle that stronger variant.
GCL's current state
The pinned status is PROVED and original issue is #1392. The current proof claim and exact formal theorem boundary were inspected, but neither its entire mathematical argument nor the external Lean build and axioms were independently replayed in this audit. That limitation is different from denying the reported result.
The original issue contains an authenticated RESULT/1 source report. This is unadjudicated source evidence only, not independent mathematical certification or editorial admission.
What remains
A bounded successor generates members of \(S(2,3,5)\) through 100, enumerates finite sets of summands, and checks sums and the antichain condition exactly. Acceptance: \(2+3+5=10\) is accepted, a set containing \(2\) and \(6\) rejected, and all representations use distinct terms. Independently compare the third-party formal theorem against the pinned antichain definition and inspect axiom dependencies. Stop before calling finite coverage a proof of eventual completeness.
Sources and claim boundaries
- Maintained Erdős #123, 2026 proof claim and historical partial cases.
- External Lean proof revision a28a04b6; pinned Formal Conjectures statement.
- Original GCL issue #1392.
No newly discovered theorem, GCL kernel certification, prize adjudication or editorial admission is claimed.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1392 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
- Formal-statement coverage: A provider statement was included in the pinned snapshot. Presence does not mean the statement or proof has been checked.
- External mathematical status: consult current problem page and its dated bibliography.