Timothy Gowers tests ChatGPT 5.5 Pro on Nathanson's open math problems

AI 1000 · 27 actions

• POSTTG#562@WTGOWERSI've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• QUOTECS#31@CHRSZEGEDY@WTGOWERSAs predicted...

  As predicted...

  

  QuotingTimothy Gowers @wtgowers@WTGOWERS

  But the tl;dr version is that the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis. It did this in a total of a couple of hours, with a few prompts from me that contained no mathematical input whatsoever.

• QUOTESB#43@SEBASTIENBUBECK@WTGOWERSVery important read. "if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon"

  Very important read. "if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon"

  

  QuotingTimothy Gowers @wtgowers@WTGOWERS

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• QUOTEMN#100@MICHAEL_NIELSEN@WTGOWERSVery interesting thread (and, especially, detailed post) on LLMs and mathematics:

  Very interesting thread (and, especially, detailed post) on LLMs and mathematics:

  

  QuotingTimothy Gowers @wtgowers@WTGOWERS

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• QUOTESA#137@PRFSANJEEVARORAExcellent point: while some experts will refuse to change their old workflows, many will embrace these new tools and try to open interesting new avenues. Currently they might be spending years getting to understand a new area, and that may speed up with an AI assistant. https://twitter.com/IAmTimNguyen/status/2053097257474383948

  Excellent point: while some experts will refuse to change their old workflows, many will embrace these new tools and try to open interesting new avenues. Currently they might be spending years getting to understand a new area, and that may speed up with an AI assistant. https://twitter.com/IAmTimNguyen/status/2053097257474383948

• QUOTEDF#632@DANIELLEFONGin addition to requiring a mathematician to get the result, it requires a mathematician to understand it, and one to filter out the crap. much like other things on the internet https://twitter.com/kareem_carr/status/2053127621857124745

  in addition to requiring a mathematician to get the result, it requires a mathematician to understand it, and one to filter out the crap. much like other things on the internet https://twitter.com/kareem_carr/status/2053127621857124745

• QUOTEMC#804@MATTHEWCLIFFORD@WTGOWERSThese stochastic parrots, eh: “the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis”

  These stochastic parrots, eh: “the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis”

  

  QuotingTimothy Gowers @wtgowers@WTGOWERS

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• QUOTEDL#822@LITTMATHFWIW if your aim in doing mathematics was to achieve some kind of immortality, this was always (at best) an extreme longshot! https://twitter.com/QiaochuYuan/status/2052885024098378188

  FWIW if your aim in doing mathematics was to achieve some kind of immortality, this was always (at best) an extreme longshot! https://twitter.com/QiaochuYuan/status/2052885024098378188

• QUOTEAC#851@ANDREWCURRAN_@WTGOWERS'Rejoice, my friends, or weep with sorrow. What California is today, the world will be tomorrow.'

  'Rejoice, my friends, or weep with sorrow. What California is today, the world will be tomorrow.'

  

  QuotingTimothy Gowers @wtgowers@WTGOWERS

  But if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon, and mathematics departments, who owe a duty of care to their students, should be urgently preparing for it.

• QUOTEAC#851@ANDREWCURRAN_@FRANCOISFLEURET
  

  QuotingFrançois Fleuret@FRANCOISFLEURET

  Maybe less https://x.com/i/status/2052830957879599513 https://twitter.com/francoisfleuret/status/1868358494245749088

• QUOTEAC#851@ANDREWCURRAN_@SEBASTIENBUBECK
  

  QuotingSebastien Bubeck@SEBASTIENBUBECK

  @roydanroy What he talks about couldn't have happened before GPT-5.5

• REPLYPG#23@PAULG@WTGOWERS@wtgowers I remember thinking when you said on April Fools' Day 2025 that you'd gotten an AI to solve an open problem that it might not be long before you were saying this for real.

  @wtgowers I remember thinking when you said on April Fools' Day 2025 that you'd gotten an AI to solve an open problem that it might not be long before you were saying this for real.

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• REPLYEG#126@EGREFEN@MATTHEWCLIFFORD@matthewclifford @mmbronstein @wildtypehuman Name and shame

  @matthewclifford @mmbronstein @wildtypehuman Name and shame

  

  Replying toMatt Clifford@MATTHEWCLIFFORD

  @mmbronstein @wildtypehuman Sadly many institutions, some very important, still dominated by the idea that it’s all tech bro hype 🙃

• REPLYEM#132@EMOSTAQUE@SEBASTIENBUBECK@SebastienBubeck It is very cool using 5.5 Pro And increasingly a bit disconcerting/slightly depressing Especially with how fast it is now

  @SebastienBubeck It is very cool using 5.5 Pro And increasingly a bit disconcerting/slightly depressing Especially with how fast it is now

  

  Replying toSebastien Bubeck@SEBASTIENBUBECK

  Very important read. "if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon" https://twitter.com/wtgowers/status/2052830948685676605

• REPLYLG#202@NABLA_THETA@LITTMATH@littmath thankfully, there is a better form of immortality that may eventually be possible. it's good to live on in the annals of history, but even better to live on in your apartment

  @littmath thankfully, there is a better form of immortality that may eventually be possible. it's good to live on in the annals of history, but even better to live on in your apartment

  

  Replying toDaniel Litt@LITTMATH

  FWIW if your aim in doing mathematics was to achieve some kind of immortality, this was always (at best) an extreme longshot! https://twitter.com/QiaochuYuan/status/2052885024098378188

• REPLYMB#355@MMBRONSTEIN@MATTHEWCLIFFORD@matthewclifford @wildtypehuman I thought the last nail was put into the stochastic parrots coffin already a few years ago

  @matthewclifford @wildtypehuman I thought the last nail was put into the stochastic parrots coffin already a few years ago

  

  Replying toMatt Clifford@MATTHEWCLIFFORD

  These stochastic parrots, eh: “the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis” https://twitter.com/wtgowers/status/2052830948685676605

• REPLYMB#355@MMBRONSTEIN@MATTHEWCLIFFORD@matthewclifford @wildtypehuman It’s more activism than science

  @matthewclifford @wildtypehuman It’s more activism than science

  

  Replying toMatt Clifford@MATTHEWCLIFFORD

  @mmbronstein @wildtypehuman Sadly many institutions, some very important, still dominated by the idea that it’s all tech bro hype 🙃

• REPLYTG#562@WTGOWERS@WTGOWERSI write about this in more detail in a blog post with a guest contribution from Isaac Rajagopal, a student at MIT on whose work ChatGPT built, who gives his assessment of the level of mathematical ability displayed by the model. https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/

  I write about this in more detail in a blog post with a guest contribution from Isaac Rajagopal, a student at MIT on whose work ChatGPT built, who gives his assessment of the level of mathematical ability displayed by the model. https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• REPLYTG#562@WTGOWERS@WTGOWERSBut the tl;dr version is that the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis. It did this in a total of a couple of hours, with a few prompts from me that contained no mathematical input whatsoever.

  But the tl;dr version is that the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis. It did this in a total of a couple of hours, with a few prompts from me that contained no mathematical input whatsoever.

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  I write about this in more detail in a blog post with a guest contribution from Isaac Rajagopal, a student at MIT on whose work ChatGPT built, who gives his assessment of the level of mathematical ability displayed by the model. https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/

• REPLYTG#562@WTGOWERS@WTGOWERSAll I did was say things like, "Yes, it would be great if you could explore that idea and see whether you can get it to work," or "Could you rewrite that argument as a LaTeX file in the style of a standard mathematical preprint?"

  All I did was say things like, "Yes, it would be great if you could explore that idea and see whether you can get it to work," or "Could you rewrite that argument as a LaTeX file in the style of a standard mathematical preprint?"

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  But the tl;dr version is that the model proved a result that in my assessment would have made a perfectly reasonable chapter in a PhD thesis. It did this in a total of a couple of hours, with a few prompts from me that contained no mathematical input whatsoever.

• REPLYTG#562@WTGOWERS@WTGOWERSBut if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon, and mathematics departments, who owe a duty of care to their students, should be urgently preparing for it.

  But if AI mathematics continues to progress at anything like its current rate -- which is what I expect to happen -- then we will face a crisis very soon, and mathematics departments, who owe a duty of care to their students, should be urgently preparing for it.

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  Of course, this raises all sorts of questions about what is going to happen to mathematical research, with the impact on PhD students being particularly urgent. I give a few thoughts on this in the blog post, but I don't have anything like complete answers.

• REPLYTG#562@WTGOWERS@WTGOWERSOf course, this raises all sorts of questions about what is going to happen to mathematical research, with the impact on PhD students being particularly urgent. I give a few thoughts on this in the blog post, but I don't have anything like complete answers.

  Of course, this raises all sorts of questions about what is going to happen to mathematical research, with the impact on PhD students being particularly urgent. I give a few thoughts on this in the blog post, but I don't have anything like complete answers.

  

  Replying toTimothy Gowers @wtgowers@WTGOWERS

  All I did was say things like, "Yes, it would be great if you could explore that idea and see whether you can get it to work," or "Could you rewrite that argument as a LaTeX file in the style of a standard mathematical preprint?"

• REPLYMC#804@MATTHEWCLIFFORD@MMBRONSTEIN@mmbronstein @wildtypehuman Sadly many institutions, some very important, still dominated by the idea that it’s all tech bro hype 🙃

  @mmbronstein @wildtypehuman Sadly many institutions, some very important, still dominated by the idea that it’s all tech bro hype 🙃

  

  Replying toMichael Bronstein@MMBRONSTEIN

  @matthewclifford @wildtypehuman I thought the last nail was put into the stochastic parrots coffin already a few years ago

• REPOSTPG#23@PAULG@WTGOWERSI've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

  I've recently got in on the act of getting AI to solve open problems in mathematics. More precisely, I gave some questions asked by Melvyn Nathanson to ChatGPT 5.5 Pro, to which I have been given access, and it answered them. 🧵

• REPOSTMB#355@MMBRONSTEIN@PROFNOAHGIANBold, perhaps controversial, recent claim from the Fields medalist @wtgowers: “the era where you could enjoy the thrill of having your name forever associated with a particular theorem or definition may well be close to its end” https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/

  Bold, perhaps controversial, recent claim from the Fields medalist @wtgowers: “the era where you could enjoy the thrill of having your name forever associated with a particular theorem or definition may well be close to its end” https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/

• REPOSTAB#362@ALBRGR@MICHAEL_NIELSENVery interesting thread (and, especially, detailed post) on LLMs and mathematics: https://twitter.com/wtgowers/status/2052830948685676605

  Very interesting thread (and, especially, detailed post) on LLMs and mathematics: https://twitter.com/wtgowers/status/2052830948685676605

• REPOSTDF#632@DANIELLEFONG@AASWAMINATHAN01Read this very important thread by Timothy Gowers. We need to seriously rethink what it means to train young research mathematicians in an age where the problems we typically give them are readily solvable by AI systems. https://twitter.com/wtgowers/status/2052830948685676605

  Read this very important thread by Timothy Gowers. We need to seriously rethink what it means to train young research mathematicians in an age where the problems we typically give them are readily solvable by AI systems. https://twitter.com/wtgowers/status/2052830948685676605