{
  "id": 826803,
  "title": "Chinese doctor stuns maths world by cracking decades-old problem using ChatGPT",
  "url": "https://urgent.news/2026/08/14/chinese-doctor-stuns-maths-world-by-cracking-decades-old-problem-826803",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-14T02:00:12.000Z",
  "source": {
    "name": "SCMP Tech",
    "slug": "scmp-tech",
    "url": "https://www.scmp.com/tech/tech-trends/article/3363966/chinese-doctor-stuns-maths-world-cracking-decades-old-problem-using-chatgpt"
  },
  "original_language": "en",
  "account": "A Chinese neurosurgeon named Jin Shanmu stunned the maths world by solving a decades-old problem using OpenAI's ChatGPT AI model. The problem, known as Crouzeix's conjecture, had baffled experts since 2004. Jin, a postdoctoral researcher at Peking Union Medical College Hospital, used a 16-hour autonomous run on GPT-5.6-Sol to crack the puzzle. The conjecture, proposed by French mathematician Michel Crouzeix, states that the norm of applying any function to a matrix is no larger than twice the function's maximum value on that matrix's numerical range. Mathematicians Alex Townsend and Anne Greenbaum from Cornell University and the University of Washington revealed that Jin had solved the problem after receiving an email from ChatGPT suggesting a solution they had just published. Both Townsend and Greenbaum, as well as Crouzeix himself, confirmed the accuracy of Jin's proof, although it has not yet undergone formal peer review. Jin's achievement is all the more remarkable given his lack of formal mathematical training, as he is an undergraduate geology major who later transitioned to medicine. This breakthrough underscores how quickly frontier AI systems are advancing and their growing role in driving mathematical discoveries.",
  "summary": "Beijing-based neurosurgeon Jin Shanmu has stunned the mathematical world by solving a two-decade-old puzzle known as Crouzeix’s conjecture, using OpenAI’s latest AI model, ChatGPT. The conjecture, proposed by French mathematician Michel Crouzeix, posits that the norm of applying any function to a matrix is no larger than twice the function’s maximum value on that matrix’s numerical range. Jin, a self-taught maths enthusiast with a background in geology and medicine, used a 16-hour autonomous run on GPT-5.6-Sol to prove the conjecture. Cornell University mathematician Alex Townsend and University of Washington professor Anne Greenbaum confirmed the proof as correct, although it has yet to undergo formal peer review. This breakthrough highlights the growing influence of frontier AI systems in driving mathematical discoveries.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "South China Morning Post",
        "title": "Chinese doctor stuns maths world by cracking decades-old problem using ChatGPT",
        "url": "https://urgent.news/2026/08/14/chinese-doctor-stuns-maths-world-by-cracking-decades-old-problem",
        "published": "2026-08-14T02:00:12.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}