Erdős Problem #52

Partial progress by Thomas Bloom, Will Sawin, Carl Schildkraut and Dmitrii Zhelezov with GPT-5.5 Pro (28 May 2026): solution to variant problem. 4 AI contributions are recorded in all.

Problem
Erdős problem #52 (number theory, additive combinatorics); current status: open
Credit
Thomas Bloom, Will Sawin, Carl Schildkraut and Dmitrii Zhelezov with GPT-5.5 Pro · AI working with humans
Checked
Listed in Tao's wiki
Timeline
First AI contribution 2026-05-28 · wiki data as of 2026-06-30
Sources
Problem page (erdosproblems.com)Tao's AI-contributions wiki
PARTIALPer Tao's AI-contributions wiki

The problem

Let AA be a finite set of integers. Is it true that for every ϵ>0\epsilon>0

max⁡(∣A+A∣,∣AA∣)≫ϵ∣A∣2−ϵ?\max( \lvert A+A\rvert,\lvert AA\rvert)\gg_\epsilon \lvert A\rvert^{2-\epsilon}?

Statement quoted from erdosproblems.com/52, maintained by Thomas Bloom. Original source: [Er77c][Er80,p.112][ErGr80][Er91][Er92c][Er95][Er97][Er97e][Va99,1.26]. Current status there: open.

AI contributions

As recorded in Terence Tao's wiki AI contributions to Erdős problems (data as of 30 Jun 2026). 🟢 full, 🟡 partial, ⚪ unverified, 🔴 incorrect.

Date Kind AI systems Humans Outcome
28 May, 2026 AI working with humans GPT-5.5 Pro Thomas Bloom, Will Sawin, Carl Schildkraut, Dmitrii Zhelezov 🟡 Solution to variant problem
28 May, 2026 AI working with humans GPT-5.5 Ingo Althöfer 🟡 Improved bound to variant problem
3 Jun, 2026 AI on its own Claude Mythos — 🟡 Solution to variant problem
4 Jun, 2026 AI working with humans GPT-5.5 Boris Alexeev ⚪ Candidate solution to variant problem

Caveats

The wiki's classifications are provisional and are not a benchmark. Many Erdős problems have never had a thorough literature review, so a result that looks new may later turn out to be known; some problems were also stated ambiguously or incorrectly. See the problem page for the full discussion.

∎