+- Scivillage.com Casual Discussion Science Forum (https://www.scivillage.com)
+-- Forum: Science (https://www.scivillage.com/forum-61.html)
+--- Forum: Computer Sci., Programming & Intelligence (https://www.scivillage.com/forum-79.html)
+--- Thread: Article The last Fields Medalist: non-AI aided human maths discoveries are obsolete (/thread-21307.html)
The last Fields Medalist: non-AI aided human maths discoveries are obsolete - C C - Sep 11, 2026
RELATED: AI solved one of math's Millennium Prize problems & everyone is angry
- - - - - - - - - - - -
The last Field's Medalist?
https://hansonwriter.substack.com/p/the-last-fields-medalist?r=u0s9r&utm_campaign=post-expanded-share&utm_medium=web
EXCERPTS: He and his co-recipients, he said, are the last Fields Medal winners who made their discoveries unaided by AI as a main research driver. AI is now so powerful that all future higher-mathematics discoveries will be made by AI, with or without human assistance.
The claim is not a flourish. It is a dating of an epoch. The medal, awarded every four years to mathematicians under forty, has always performed a double function: it certifies past work and anoints future leadership. Tsimerman’s remark converts the second function into an epitaph. After this cohort, the human name on the medal will require a footnote.
[...] Tsimerman said several times that mathematicians attack problems, or seek discoveries, largely in areas where they think they can make some progress. He used the phrase “low-hanging fruit.” He explained that he set aside various problems that interested him in order to focus for two years on problems that, if solved, gave him a good chance of winning the Fields. He thought it would take two years. He thought he had a reasonable chance. If he failed, the cost was only two years.
Elsewhere he called the habit of addressing problems one thinks solvable within a reasonable period “pragmatic.” Low-hanging fruit, in his usage, are solutions close to what we already know.
At one stage he said the quiet part aloud: AI has no such constraint. It goes about solving problems irrespective of the prospects of success. It is not worried about failure. It is not worried what the tenure committee or the VC will think.
Hence the radical thesis: more problems will be solved with AI, but those solutions will sit further from our current understanding. [...] The professional instinct he describes is a form of prudence: Aim too far and you vanish. Aim too near and you merely rearrange what is already known. The good practitioner works the edge. AI, as medium, does not inhabit that edge. It does not feel the cost of missing it. The frontier therefore ceases to be a humanly calibrated line and becomes a discontinuous jump... (MORE - missing details)