(no subject)
Sep. 5th, 2026 02:04 pmAlternate history from Terry:
Aug 10, 2005: Goldston, Pintz and Yildirim develop a method that can obtain a bound of 16 on gaps between primes assuming the (still unproven) Elliott-Halberstam conjecture, and a "near miss" towards any finite bound at all without that conjecture. They also observe that any improvement over the Bombieri-Vinogradov equidistribution theorem for primes would produce *some* finite bound for gaps between primes.
Late August, 2005: Multiple AI companies race to release unconditional versions of the GPY theorem, with various bounds.
Sep 2005: After dozens of such releases, and millions of dollars of compute collectively spent, the bound stabilizes in the low triple digits. Progress stalls, and the AI companies move their attention on to the next benchmark.
2005-: The problem is now considered "saturated", requiring enormous amounts of compute to make further progress. Mathematicians now also move their attention to other problems in the field
2005-2010 The bound occasionally drops over the years, as bored individuals with extra compute credits point their latest AI models at the problem to squeeze a bit more out of the proof techniques. However, these improvements gather almost no attention, either within the professional mathematics community or on social media.
~2010: One of the few mathematicians remaining in the area sifts through the hundreds of pages generated by the frenzy of 2005, and eventually teases out the idea of the Maynard sieve that was buried in some of the AI-generated results. However, this discovery is kept private, as by this time any release of a new mathematical insight inevitably leads to a further AI-generated frenzy of "benchmark optimization". This mathematician instead secretly works to try to use this idea to solve the full twin prime conjecture.
~2012: Even with the latest AI assistance, said mathematician fails to resolve the full conjecture, and abandons the effort without publishing the partial results. (In any event, the prestige value of solving the twin prime conjecture had become greatly diminished after years of AI-generated solutions to other long-standing mathematical problems.)
2013: Zhang continues to work at a sandwich shop. Maynard switches out of analytic number theory to a different topic of study.
Aug 10, 2005: Goldston, Pintz and Yildirim develop a method that can obtain a bound of 16 on gaps between primes assuming the (still unproven) Elliott-Halberstam conjecture, and a "near miss" towards any finite bound at all without that conjecture. They also observe that any improvement over the Bombieri-Vinogradov equidistribution theorem for primes would produce *some* finite bound for gaps between primes.
Late August, 2005: Multiple AI companies race to release unconditional versions of the GPY theorem, with various bounds.
Sep 2005: After dozens of such releases, and millions of dollars of compute collectively spent, the bound stabilizes in the low triple digits. Progress stalls, and the AI companies move their attention on to the next benchmark.
2005-: The problem is now considered "saturated", requiring enormous amounts of compute to make further progress. Mathematicians now also move their attention to other problems in the field
2005-2010 The bound occasionally drops over the years, as bored individuals with extra compute credits point their latest AI models at the problem to squeeze a bit more out of the proof techniques. However, these improvements gather almost no attention, either within the professional mathematics community or on social media.
~2010: One of the few mathematicians remaining in the area sifts through the hundreds of pages generated by the frenzy of 2005, and eventually teases out the idea of the Maynard sieve that was buried in some of the AI-generated results. However, this discovery is kept private, as by this time any release of a new mathematical insight inevitably leads to a further AI-generated frenzy of "benchmark optimization". This mathematician instead secretly works to try to use this idea to solve the full twin prime conjecture.
~2012: Even with the latest AI assistance, said mathematician fails to resolve the full conjecture, and abandons the effort without publishing the partial results. (In any event, the prestige value of solving the twin prime conjecture had become greatly diminished after years of AI-generated solutions to other long-standing mathematical problems.)
2013: Zhang continues to work at a sandwich shop. Maynard switches out of analytic number theory to a different topic of study.