Contested
Erdős problem #358: argument judged by the wiki to have major gaps (GPT-5.2 Thinking)
The gist
According to the wiki of AI contributions to Erdős problems, Nat Sothanaphan reportedly put forward with GPT-5.2 Thinking, between February 13 and 20, 2026, an argument for problem #358.
DiscoverIA status on October 12, 2026: contested, meaning sourced public objection, with no established refutation.
AI draft, not reviewedWhat has not been reviewed?
Text written by an AI (DiscoverIA pipeline), on . The status and its sources come from the registry; only the explanatory text has not been reviewed. A reported error is corrected, then the text is reviewed.
Cite this entry
The status has changed 1 times since Oct 12, 2026: a citation always points to a dated snapshot. Cite an earlier snapshot
Cite snapshot 2, Oct 12, 2026
APA citation with the dated link
BibTeX @misc, key discoveria-erdos-358-gpt-5-2-thinking-1d-r2
@misc{discoveria-erdos-358-gpt-5-2-thinking-1d-r2,
author = {{DiscoverIA}},
title = {Erdős problem #358: argument judged by the wiki to have major gaps (GPT-5.2 Thinking)},
howpublished = {DiscoverIA registry, entry erdos-358-gpt-5-2-thinking-1d},
note = {Status “Contested”, snapshot 2, recorded on October 12, 2026. New result},
year = {2026},
url = {https://discoveria.ai/entry/erdos-358-gpt-5-2-thinking-1d#releve-2},
urldate = {2026-10-12}
}Dated link opens this snapshot in the history
To understand ≈ 30 s reading time, estimated
What it changes
Problem #358 is a number theory question touching on additive bases (sets of integers such that every sufficiently large integer is a sum of a bounded number of their elements). The wiki describes the argument as an “Argument with major gaps made” and flags it in red: in its view it is not a solution, and the entry records an attempt made by a human with an AI.
What remains unclear
The wiki has been frozen since June 30, 2026, so anything published since then would not appear there. The sources read do not say where the gaps lie, nor whether the problem's current state on erdosproblems.com, recorded on October 12, 2026, is due to this argument.
For experts
Sources, method, key facts, status history, report an error
Status
Contested
Recorded on · github.com · wiki
Outside the verification path
What this status means
Source reading : Source read word for word (level 3 of 3)
Sourced public objection, with no established refutation.
It stays “Contested” until a refutation is established. It would become “Withdrawn” if its author withdrew it or the reference source marked it incorrect.
Dated archive: status recorded on the wiki frozen on Jun 30, 2026
Sources
- Main sourcegithub.com · wikipage read word for word · consulted on Oct 12, 2026
- Reference listerdosproblems.comlink only, reading not recorded
- Corroborationterrytao.wordpress.compage read word for word · consulted on Oct 17, 2026 ·
We have created a github repository to coordinate this process
- Repositorygithub.com · CONTRIBUTING.mdpage read word for word · consulted on Oct 17, 2026 ·
this status is an unofficial crowdsourced classification of the problem
Method
Source: the “AI contributions to Erdős problems” wiki of the teorth/erdosproblems repository (a repository presented by Terence Tao in his blog post of August 31, 2025); frozen on June 30, 2026 (commit c8ad430, “freeze”) and read on October 12, 2026 in the git repository. Category 1(d), “AI collaborating with humans” (significant human involvement). Humans named by the wiki: Nat Sothanaphan; AI system: GPT-5.2 Thinking. Dates given by the wiki: “13-20 Feb, 2026”. Exact text: “Argument with major gaps made”; red flag (“incorrect work”, according to the wiki's legend). Literature cited by the wiki: none. Status of the problem in the teorth/erdosproblems community database (data/problems.yaml, unofficial classification, snapshot of October 12, 2026 with no commit identified, current state, not that of June 2026): proved, answered in the affirmative (informal_status=proved); a solution has been formalised in Lean (formal_status=Lean). This status concerns the problem, not this contribution. Problem page not read by DiscoverIA: https://www.erdosproblems.com/358. The DiscoverIA registry also holds an entry that groups several problems, this one included; the two entries may overlap. DiscoverIA status: contested (a sourced public objection exists, with no established refutation). Recorded on October 12, 2026. History: 2 dated observations, oldest first.
Limits
DiscoverIA has not checked this contribution itself.
Key facts
- Claimed by
- Nat Sothanaphan with GPT-5.2 Thinking, according to the AI-Erdős wiki
announced on Feb 13, 2026 - Nature
- new result
- Field
- Mathematics (Erdős problem #358 (number theory ; additive basis ; primes))
- Wiki category
- AI collaborating with humans (IA en collaboration avec des humains ; implication humaine significative)
- Wiki verdict
- « Argument with major gaps made » : travail incorrect (légende du wiki)
- Checked by
- wiki maintainers (red flag)
- Sources
- 4 : 3 read word for word, 1 link only
- This entry
- AI draft, not reviewed
Status history
Most recent first; each observation is numbered in the order it was recorded.
- 02
WithdrawnbecomesContestedCurrent status
reclassified from withdrawn to contested based on the wiki's text
Source: github.comread word for word
- 01
First observation:Withdrawn
marked incorrect by the wiki (red flag, “Incorrect work”)
Source: github.comread word for word
An observation is added; none is ever erased.
Cite an earlier snapshot
These snapshots no longer give the current status. Cite them only to date a past state.
Cite snapshot 1, Oct 12, 2026
Withdrawn status superseded on Oct 12, 2026
marked incorrect by the wiki (red flag, “Incorrect work”)
APA citation with the dated link
BibTeX @misc, key discoveria-erdos-358-gpt-5-2-thinking-1d-r1
@misc{discoveria-erdos-358-gpt-5-2-thinking-1d-r1,
author = {{DiscoverIA}},
title = {Erdős problem #358: argument judged by the wiki to have major gaps (GPT-5.2 Thinking)},
howpublished = {DiscoverIA registry, entry erdos-358-gpt-5-2-thinking-1d},
note = {Status “Withdrawn”, snapshot 1, recorded on October 12, 2026. New result},
year = {2026},
url = {https://discoveria.ai/entry/erdos-358-gpt-5-2-thinking-1d#releve-1},
urldate = {2026-10-12}
}Dated link opens this snapshot in the history
In the same set
AI contributions to Erdős problems (wiki by T. Tao, frozen on June 30, 2026) · entry 134 of 521, by problem number
- Not evaluated : Erdős problem #354: partial results found in the literature according to the wiki (ChatGPT Deep research)
- Not evaluated : Erdős problem #355: formalisation by Aristotle, Gemini 3 (proof: van Doorn and Kovač (2025) (almost the entirety))
- Contested : Erdős problem #358: argument judged by the wiki to have major gaps (GPT-5.2 Pro)
- Contested : Erdős problem #358: argument judged by the wiki to have major gaps (GPT-5.2 Thinking)
- Not evaluated : Erdős problem #363: formalisation by Aristotle (proof: Ulas (2005))
- Not evaluated : Erdős problem #366: conditional partial result according to the wiki (Automated research pipeline)
- Not evaluated : Erdős problem #367: formalisation by Aristotle (proof: van Doorn and Tao (2025))
Report an error or suggest a source
Thank you, your message was sent.
Sending failed. Try again, or write to contact@discoveria.ai.
Is your result listed?What each status requires, correcting this entry, replying to it.