AI contributions to ErdΕ‘s problems

12 min read Original article β†—

This page collects the various ways in which AI tools have contributed to the understanding of ErdΕ‘s problems. Note that a single problem may appear multiple times in these lists. For further discussion of these contributions, see this forum.

Legend

  • 🟒 : full solution to the problem
  • 🟑 : partial solution to the problem
  • πŸ”΄ : failure to make progress on the problem

Disclaimers

One should interpret the contributions listed here with the following caveats in mind.

  • ErdΕ‘s problems vary widely in difficulty (by several orders of magnitude), with a core of very interesting, but extremely difficult problems at one end of the spectrum, and a "long tail" of under-explored problems at the other, many of which are "low hanging fruit" that are very suitable for being attacked by current AI tools. Unfortunately, it is hard to tell in advance which category a given problem falls into, short of an expert literature review. (However, if an ErdΕ‘s problem is only stated once in the literature, and there is scant record of any followup work on the problem, this suggests that the problem may be of the second category.) In particular, I would advise against direct comparison of raw counts of problems solved by one methodology or another, as this may not be an apples-to-apples comparison.
  • Many of the problems on the site currently lack a thorough literature review, and in the absence of such a review, any designation of a problem as "open" should be viewed as provisional. In particular, it has already happened multiple times that an AI tool has managed to solve a problem listed as open on the site, only to find shortly afterwards that the problem was in fact solved in the literature (though sometimes by a slightly different method); see Section 2 below.
  • The reporting of these AI tools on this page is highly incomplete, particularly with regards to negative results. One should keep selection bias in mind before drawing any conclusions about the success rate of these tools.
  • In a few cases ErdΕ‘s stated the problem incorrectly, allowing for the problem as literally stated to be solved on a technicality, even though this was not the one ErdΕ‘s intended. Working out what ErdΕ‘s actually intended in such cases requires a somewhat subjective contextual analysis guided by expertise in the subject.
  • For additional guarantees of correctness, it is considered good practice to formalize AI-generated proofs in a proof assistant language such as Lean. However, there are still "exploits" available in such guarantees, for instance if the formal proof introduces additional axioms, or if there is a misformalization of the problem statement that may exploit some quirk of the math library or syntax used in the formalization. An AI-generated formal proof that is suspiciously short and elementary (or suspiciously verbose and trivial) is particularly subject to this issue, particularly in the hands of a user without extensive experience formalizing proofs by hand.

I just saw an exciting announcement on social media about AI being used to solve an ErdΕ‘s problem. Why can't I find out about it on this page?

If the social media post is very recent, then most likely the claim is undergoing an expedited, but not instantaneous, peer review over at the ErdΕ‘s problem web site. The discussion page for the ErdΕ‘s problem in question is most likely the best place to get the most recent updates, and in particular to perform the due diligence needed to take into account the caveats listed above. Once the discussion on that page has reached consensus, this contribution will most likely appear on this site; but until then, I recommend taking any bold claims about this achievement with a grain of salt, and check back in a couple hours or days for a more accurate update.

1. AI-generated solutions, partial solutions, or negative results for previously open problems

Problem AI tools used Date Outcome
[36] AlphaEvolve 3 Nov, 2025 🟑 Slight improvement to past construction
[52] AlphaEvolve 3 Nov, 2025 πŸ”΄ Did not match past constructions
[64] AlphaEvolve 3 Nov, 2025 No counterexample found
[67] AlphaEvolve 3 Nov, 2025 πŸ”΄ Did not match past constructions
[106] AlphaEvolve 3 Nov, 2025 Matched past construction
[124] Aristotle 29 Nov, 2025 🟑 Partial result
[391] AlphaEvolve 3 Nov, 2025 πŸ”΄ Did not match past constructions
[493] AlphaEvolve 3 Nov, 2025 No counterexample found
[507] AlphaEvolve 3 Nov, 2025 🟑 Surpassed some past constructions
[728] Aristotle, ChatGPT 5.2 Pro 6 Jan, 2026 🟒 Full solution (Lean)
[1097] AlphaEvolve 3 Nov, 2025 🟑 Slight improvement to past construction

2. Fully AI-generated solutions to problems thought open, but for which an earlier human solution was subsequently found

Problem AI tools used Date Outcome Literature result Literature result found on Similar proofs?
[333] ChatGPT 5.2 Pro, Claude Opus 4.5 25 Dec, 2025 🟒 Full solution 🟒 ErdΕ‘s and Newman (1977) 25 Dec, 2025 Yes
[897] Archivara 26 Dec, 2025 🟒 Full solution 🟒 Wirsing (1981) 26 Dec, 2025 Yes
[1026] Aristotle 7 Dec, 2025 🟒 Full solution 🟒 Tidor, Wang, and Yang (2016) 8 Dec, 2025 Only after applying an argument from Seidenberg (1959)
[1077] Aristotle 24 Dec, 2025 🟑 Trivial counterexample to previous formulation of problem 🟒 Jiang and Longbrake (2025) 28 Dec, 2025 No

3. AI-generated tools applied to problems previously known to be solved (or partially solved)

Problem AI tools used Date Literature result Outcome
[43] Aristotle 4 Dec, 2025 🟑 Barreto (2025) 🟑 New proof of partial result
[124] Gemini DeepThink 30 Nov, 2025 🟑 Aristotle (2025) πŸ”΄ Did not reproduce a known partial result despite hints
[198] AlphaProof 2025 🟒 Baumgartner (1975) 🟒 New proof found
[264] Aristotle 18 Dec, 2025 🟑 Kovač and Tao (2024) 🟑 New proof of partial result; πŸ”΄ no progress on remaining open problems
[392] GPT-5.2 Pro, Aristotle 31 Dec, 2025 🟒 ErdΕ‘s and Graham (unpublished, 1980), Cambie (2025) Unable to reconstruct proof
[488] Aristotle 27 Nov, 2025 🟑 Cambie (2025) 🟑 New disproof of alternate version
[493] SeedProver, Aristotle, ChatGPT 2025 🟒 Seamans (2025) 🟒 New or existing proof found
[871] Claude Opus 4.5 5 Jan, 2026 🟑 ErdΕ‘s and Nathanson (1989) 🟒 Existing partial result upgraded to full solution
[942] Gemini 23 Nov, 2025 🟑 ErdΕ‘s (unpublished, 1976) 🟑 New proof found
[958] Aristotle 27 Dec, 2025 🟒 Clemen, Dumitrescu, and Liu (2025) 🟒 New proof found
[1043] Aristotle 28 Dec, 2025 🟒 Pommerenke (1961) 🟒 New proof found
[1043] ChatGPT 30 Dec, 2025 🟒 Pommerenke (1961), Aristotle 🟒 New proof found
[1095] Aristotle 30 Dec, 2025 🟑 Ecklund, ErdΕ‘s, and Selfridge (1975) 🟑 New proof of (slightly weaker) partial result found; πŸ”΄ no progress on remaining open problems

4. Solutions generated by humans in collaboration with AI

Problem Human collaborators AI tools used Date Outcome
[367] Boris Alexeev, Wouter van Doorn, Terence Tao Gemini Deepthink, Aristotle 20-22 Nov, 2025 🟑 Partial result
[848] Mehtaab Sawhney, Mark Sellke GPT-5 12 Oct-20 Nov, 2025 🟒 Full solution
[1026] Boris Alexeev, Stijn Cambie, Terence Tao, Lawrence Wu ChatGPT, Aristotle, AlphaEvolve, Gemini 8 Dec, 2025 🟒 Full solution
[1038] Junnosuke Koizumi, jspier, Nat Sothanaphan, Terence Tao ChatGPT, AlphaEvolve In progress 🟑 Partial result

5. Problems with an AI-powered literature review

Problem AI tools used Review performed on Outcome
[35] GPT-5 13 Oct, 2025 🟑 Partial results found
[66] GPT-5 13 Oct, 2025 🟑 Partial results found
[94] GPT-5 2 Nov, 2025 🟒 Full solution found
[124] ChatGPT DeepResearch, Gemini DeepResearch 30 Nov, 2025 No significant results found
[167] GPT-5 12 Oct, 2025 🟑 Partial results found
[188] GPT-5 13 Oct, 2025 🟑 Partial results found
[203] ChatGPT DeepResearch, Gemini DeepResearch 19 Oct, 2025 No significant results found
[223] GPT-5 13 Oct, 2025 🟒 Full solution found
[248] Gemini DeepResearch 19 Oct, 2025 No significant results found
[330] ChatGPT DeepResearch, Gemini DeepResearch, Claude 19 Dec, 2025 πŸ”΄ Partial results found (with inaccuracies); did not find literature proof
[333] ChatGPT 5.2 Pro 25 Dec, 2025 πŸ”΄ Incorrect proof claimed; did not find literature proof
[334] Gemini DeepResearch 19 Nov, 2025 No significant results found
[339] GPT-5 11 Oct, 2025 🟒 Full solution found
[347] ChatGPT DeepResearch 25 Oct, 2025 🟒 Full solution found
[354] ChatGPT DeepResearch 19 Oct, 2025 🟑 Partial results found
[367] ChatGPT DeepResearch, Gemini DeepResearch 22 Nov, 2025 πŸ”΄ No significant results found; did not find ErdΕ‘s problems community proof
[370] ChatGPT DeepReseach, Gemini, Gemini DeepResearch 17 Oct, 2025 🟑 Problem found to be misstated; solutions to variants found
[387] ChatGPT DeepResearch 1 Nov, 2025 🟑 Partial results found
[421] ChatGPT DeepResearch, Gemini DeepResearch 18 Oct, 2025 No significant results found
[434] GPT-5 29 Oct, 2025 πŸ”΄ Literature proof not found
[481] ChatGPT DeepResearch, Gemini DeepResearch 1 Dec, 2025 🟑 Unwittingly reproduced existing proof
[481] ChatGPT 3 Dec, 2025 🟒 Full solution found
[494] GPT-5 13 Oct, 2025 🟒 Full solution found
[515] GPT-5 15 Oct, 2025 🟒 Full solution found
[516] ChatGPT DeepResearch, Claude, Gemini DeepResearch 28 Dec, 2025 🟒 Problem found to be misstated; full solution to actual problem found
[524] ChatGPT 5.2 Pro 27 Dec, 2025 🟑 Partial results found
[559] ChatGPT DeepResearch, Gemini DeepResearch 26 Oct, 2025 🟑 Partial results found
[621] GPT-5 13 Oct, 2025 🟒 Full solution found
[645] GPT-5 20 Oct, 2025 🟒 Full solution found
[686] ChatGPT, Gemini DeepThink, Gemini DeepResearch 18 Oct, 2025 No significant results found
[689] ChatGPT DeepResearch, Gemini DeepResearch 29 Oct, 2025 No significant results found
[700] ChatGPT DeepResearch 17 Dec, 2025 No significant results found
[707] ChatGPT 22 Oct, 2025 πŸ”΄ Literature proof not found
[728] ChatGPT DeepResearch 15 Dec, 2025 No significant results found
[737] GPT-5 30 Sep, 2025 🟒 Full solution found
[750] GPT-5 13 Oct, 2025 🟑 Partial results found
[786] ChatGPT DeepResearch, Gemini DeepResearch 18 Oct, 2025 No significant results found
[788] GPT-5 14 Oct, 2025 🟑 Partial results found
[793] ChatGPT DeepResearch 30 Nov, 2025 🟑 Partial results found
[811] GPT-5 13 Oct, 2025 🟑 Partial results found
[822] GPT-5 13 Oct, 2025 🟒 Full solution found
[827] GPT-5 14 Oct, 2025 🟑 Partial results found
[829] GPT-5 13 Oct, 2025 🟑 Partial results found
[871] ChatGPT DeepResearch 6 Dec, 2025 🟑 Partial results found
[903] GPT-5 14 Oct, 2025 🟒 Full solution found
[906] ChatGPT DeepReseach, Gemini DeepResearch 17 Oct, 2025 No significant results found
[915] ChatGPT DeepResearch 26 Oct, 2025 🟑 Partial results found
[940] ChatGPT DeepResearch 24 Oct, 2025 No significant results found
[942] ChatGPT DeepResearch 23 Nov, 2025 🟑 Partial results found
[965] ChatGPT 2 Jan, 2026 🟒 Full solution found
[967] GPT-5 29 Oct, 2025 No significant results found
[990] ChatGPT DeepResearch, Gemini DeepResearch 20 Oct, 2025 🟑 Partial results found
[1002] ChatGPT DeepResearch, Gemini DeepResearch 3 Nov, 2025 No significant results found
[1008] GPT-5 29 Sep, 2025 🟒 Full solution found
[1011] GPT-5 13 Oct, 2025 🟑 Partial results found
[1016] ChatGPT DeepResearch, Gemini DeepResearch 18 Oct, 2025 🟑 Partial results found
[1019] ChatGPT DeepResearch, Gemini DeepResearch 18 Oct, 2025 🟑 Solution claim found, but precise citation not found
[1022] ChatGPT DeepResearch, Gemini DeepResearch 4 Dec, 2025 No significant results found
[1038] Gemini DeepResearch 16 Nov, 2025 No significant results found
[1041] GPT-5 16 Nov, 2025 No significant results found
[1043] GPT-5 12 Oct, 2025 🟒 Full solution found
[1079] GPT-5 13 Oct, 2025 🟒 Full solution found
[1099] ChatGPT DeepResearch, Gemini, Gemini DeepResearch 19 Oct, 2025 🟒 Full solution found
[1124] ChatGPT 31 Dec, 2025 🟒 Full solution to proposed variant found

6. Proofs that were formalized by AI

Problem Proof to be formalized by AI tools used to formalize Formalized on Other known proofs
[26] 🟒 Ruzsa Aristotle 28 Dec, 2025
[31] 🟒 Lorentz (1954), Wouter van Doorn (2025) ChatGPT, Aristotle 24 Nov, 2025
[43] 🟑 Barreto (2025) Claude, Aristotle 21 Dec, 2025
[56] 🟒 Ahlswede-Khachatrian (1995) ChatGPT, Aristotle 25 Nov, 2025
[105] 🟒 Xichuan (2025) ChatGPT Pro, Aristotle 17 Nov, 2025
[106] 🟑 Baek, Koizumi, and Ueoro (2024) Aristotle 10 Dec, 2025
[189] 🟒 Kovač (2023) Gemini 3 Pro, Aristotle 17 Dec, 2025
[198] 🟒 Baumgartner (1975) ChatGPT, Aristotle 24 Nov, 2025
[226] 🟒 Sato and Rankin (1974) Aristotle 29 Dec, 2025 🟒 Barth and Schneider (1970)
[229] 🟒 Barth and Schneider (1972) Aristotle 28 Dec, 2025
[246] 🟒 Birch (1959) Aristotle 28 Dec, 2025
[303] 🟒 Brown and Râdl (1991) SeedProver 21 Dec, 2025
[337] 🟒 Ruzsa and TurjÑnyi (1985) Aristotle 10 Dec, 2025
[350] 🟒 Ryavec (1974) ChatGPT, Aristotle 25 Nov, 2025
[367] 🟑 ErdΕ‘s problems community (2025) Aristotle 22 Nov, 2025
[370] 🟒 Steinerberger (2025) ChatGPT, Aristotle 24 Nov, 2025
[418] 🟒 Browkin and Schinzel (1995) ChatGPT, Aristotle 22 Nov, 2025
[480] 🟒 Chung and Graham (1984) Aristotle 28 Nov, 2025
[481] 🟒 Barreto (2025) Claude, Aristotle 1 Dec, 2025 🟒 Klarner (1982)
[499] 🟒 Marcus and Minc (1962) Aristotle 29 Nov, 2025
[541] 🟒 Grynkiewicz (2011) ChatGPT, Aristotle 30 Dec, 2025 🟒 Gao, Hamidoune, and Wang (2010)
[613] 🟒 Pikhurko (2001) ChatGPT Pro 4 Nov, 2025
[645] 🟒 Brown and Landman (1999) ChatGPT, Aristotle 23 Nov, 2025
[707] 🟒 Hall (1947) ChatGPT 23 Nov, 2025 🟒 Alexeev and Mixon (2025)
[897] 🟒 ErdΕ‘s and Wirsing (1975) Aristotle 27 Nov, 2025
[958] 🟒 Clemen, Dumitrescu, and Liu (2025) Bytedance Seed AI4Math 2025
[967] 🟒 Yip (2025) Aristotle 19 Dec, 2025
[1000] 🟒 Haight ChatGPT, Aristotle 28 Dec, 2025
[1034] 🟒 Ma-Tang (2025) Aristotle 4 Dec, 2025
[1080] 🟒 De Caen and Székely (1992) Aristotle 28 Dec, 2025

7. Human solutions with some secondary AI tool use

Problem Result Date AI tools used Nature of tool use
[69], [248], [946] 🟒🟒🟑 Tao and Teravainen 1 Dec, 2025 ChatGPT Numerics and image generation
[114] 🟑 Tao 13 Dec, 2025 Gemini, AlphaEvolve Initial numerics, code and image generation
[682] 🟒 Gafni and Tao 8 Aug, 2025 ChatGPT Initial code generation

8. Pending assessment

Problem AI tools used Date Status
[871] Claude Opus 4.5, Gemini 3 Pro 5 Jan, 2026 Obtained formal proof, but how much to credit AI parts is still unclear