OpenAI's GPT-5.2 Pro has aided in solving another Erdős problem. Neel Somani utilized the AI model to crack the number theory behind Erdős problem #281. Mathematician Tao Zhelun described this as "perhaps the clearest example yet of an AI solving an unsolved mathematical problem." Although earlier proofs may have influenced the model's answer, Tao Zhelun confirmed that GPT-5.2 Pro's proof is "quite distinct."

However, Tao Zhelun cautions about a bias in public perception regarding AI capabilities. Negative results are seldom publicized, while positive outcomes spread rapidly. A new database developed by Paata Ivanisvili and Mehmet Mars Seven tracks AI attempts on Erdős problems, revealing an actual success rate of only 1% to 2%, primarily concentrated on simpler issues.

Nevertheless, despite Tao Zhelun's belief that medium-difficulty Erdős problems remain challenging, AI is still a valuable tool. The first autonomously solved Erdős problem confirmed by Tao Zhelun dates back to January 4, 2026.