{
  "id": 7604954,
  "title": "An Existential Crisis in Math",
  "url": "https://urgent.news/2026/09/15/an-existential-crisis-in-math",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-15T18:41:00.000Z",
  "source": {
    "name": "The Atlantic",
    "slug": "the-atlantic",
    "url": "https://www.theatlantic.com/technology/2026/09/math-crisis-openai-millennium-prize/688631/"
  },
  "original_language": "en",
  "account": "In late July, renowned mathematician Terence Tao warned of a \"turbulent period\" for his field as AI grows increasingly proficient with mathematical concepts. At the International Congress of Mathematicians, Tao expressed concern that AI is challenging traditional mathematical values and practices.\n\nEarlier that week, Anthropic researcher Claude disproved an 87-year-old math conjecture, yet many mathematicians at the conference remained unaware of the topic. Some expressed surprise at journalists asking about AI's impact on their work.\n\nOpenAI announced that its AI agents solved the Navier–Stokes problem, one of the seven Millennium Prize problems, in less than four days. This breakthrough raised questions about the role of AI in mathematics, leading to an existential reckoning for the discipline. Mathematicians realized the immense power AI now wields over the field.\n\nThe OpenAI solution was based on the extensive work of previous mathematicians, but it raised concerns that AI companies might be scooping ideas from mathematicians. Max Weinreich, a math professor at CUNY Baruch, noted that AI had \"ravaged\" the field.\n\nThe Navier–Stokes problem, which has stumped mathematicians since the 1930s, illustrates the potential disconnect between AI solutions and their practical value. Although the problem deals with fluid dynamics—a critical area of study for various industries—the actual result may not directly aid further discoveries or human understanding of mathematics.\n\nDespite the potential of AI-generated proofs, there is a risk that mathematicians may struggle to understand and apply these results. Even OpenAI's proof, which took 166 pages to complete, is described as \"almost incomprehensible.\" As the field adapts to AI's growing influence, mathematicians must find ways to maintain their understanding of the proofs and ensure that AI's power serves to enhance rather than overpower the discipline.",
  "summary": "AI just solved one of the world’s hardest math problems.",
  "key_points": [
    "Renowned mathematician Terence Tao warns of AI's turbulent impact on mathematics",
    "Anthropic's AI disproves 87-year-old math conjecture, unnoticed by many mathematicians",
    "OpenAI's AI solves Navier–Stokes problem, sparking existential reckoning in the field"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "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."
}