Erdős Problem #203
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: primes, covering systems.
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
Does there exist an integer \(m\ge1\) relatively prime to six such that every integer
is composite? The exponents \(k,\ell\) vary independently, including zero. The requirement \(\gcd(m,6)=1\) is not decorative: admitting divisibility by two or three changes the residue and parity structure.
This generalizes the idea of a Sierpiński number, an odd integer \(m\) for which \(2^km+1\) is composite for every \(k\ge1\). The two-base question needs one common integer \(m\) defeating every pair of exponents.
A small example
For \(m=5\), the choice \(k=\ell=0\) gives \(m+1=6\), composite, but the choice \(k=1,\ell=0\) yields \(11\), prime. Thus \(m=5\) fails. For \(m=1\), the zero-exponent term equals two, prime, so it also fails.
An experiment that checks finitely many pairs \((k,\ell)\) can rule out a candidate by finding one prime, but cannot certify a candidate that survives the test: infinitely many pairs remain. A covering-system argument would have to cover all exponent pairs and ensure that the designated divisor is proper for each value.
What is known
The question remains OPEN, according to Erdős #203, last edited January 2026. The original question is from Erdős and Graham (1980), page 27. Ordinary Sierpiński examples solve a one-base version only.
The catalogue also records a broad variant with arbitrary fixed distinct primes \(p_1,\ldots,p_r\) and a variant where the primes \(q_i\) range through those congruent to one modulo four. A current comment observes a trivial \(m=1\) case for the latter unless additional restrictions are imposed. This does not resolve the displayed two-base question, for which the zero-exponent term \(m+1\) remains binding.
GCL's current state
Pinned/live status: OPEN. The authenticated source issue is #1472. Its protected Formal Conjectures provider contains a statement file for #203, but its exact representation of \(k,\ell\ge0\), primality versus compositeness and \(\gcd(m,6)=1\) has not been replayed. Statement presence is not a proof.
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 checks all \(1\le m\le1000\) coprime to six against exponent pairs \(0\le k,\ell\le8\), using an exact primality test and reporting a prime witness for each eliminated candidate. Acceptance: \(m=1\) fails at \((0,0)\) and \(m=5\) fails at \((1,0)\); any surviving candidate is labelled finite-window survivor, never a solution. A separate proof task would develop a genuine covering of all exponent pairs.
Sources and claim boundaries
- Erdős #203, original Erdős–Graham 1980 and variants.
- GCL issue #1472, locked statement-provenance requirements.
No universal composite construction, independent formal proof or editorial admission is claimed.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1472 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: PROTECTED_READY_BACKLOG.
- 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.