• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Report

    Fields Medalists’ AI declaration sparks debate over proofs and understanding

    Posts say 25 Fields Medal winners signed a declaration warning of a mismatch between AI companies and mathematics. Supporters stress understanding and scientific communities; critics say AI proofs can still yield insights.

    Yann LeCunYL
    Sasha RushSR
    Nando de FreitasND
    201 Sources, ,

    TLDR

    September 11 posts describe a declaration signed by 25 Fields Medal winners warning of what they see as a “severe misalignment” between AI companies and the mathematics community. A passage quoted alongside the announcement argues that producing answers can become disconnected from the training that develops understanding and new ideas.

    Supportive reactions emphasize nurturing scientific communities and intuition, rather than focusing only on results. Critics argue that AI proofs can be studied and built upon—and that solving a problem need not end exploration, because understanding the solution can lead to new methods and insights.

    Combined views

    10.8M

    201 Sources, first seen 26d ago

    Combined views

    10.8M

    201 Sources, first seen 26d ago

    51.6K likes
    26d ago
    first seen 26d ago
    51.6K likes
    3.9K comments
    9.6K saves
    6.1K reposts
    3.9K comments
    9.6K saves
    6.1K reposts

    Sentiment

    Positive20.4%79.6%Negative

    Summary

    Negative replies attacked warnings about AI in mathematics and Fields Medalist concerns as entitled or self-serving, while positive accounts praised the alerts and suggestions for AI-resistant careers like watchmaking.

    Based on 3041 sentiment-bearing replies from 2156 accounts across 14 conversations.

    Sentiment

    Positive20.4%79.6%Negative

    Summary

    Negative replies attacked warnings about AI in mathematics and Fields Medalist concerns as entitled or self-serving, while positive accounts praised the alerts and suggestions for AI-resistant careers like watchmaking.

    Based on 3041 sentiment-bearing replies from 2156 accounts across 14 conversations.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    201 Sources

    Shital Shah@sytelusSame thing has *already* happened in software engineering. They took it with grace, excitement and without whining about why their role is still important. No software engineer has ever argued that if non-professionals starts generating code for the same thing they are working on then it takes away their ability to get insights. No software engineer has ever created a petition to cancel undergraduate AI hackathons.26d
    Charles@CharlesDardamanThis is a commonly held view by a lot of people who only really studied computers or math. Since those subjects were difficult and AI can do them very well, they assume that all other “easier” fields will be just as easy to master. But they fundamentally misunderstand the difficulty of mastering other fields, especially those that interact with the real world.26d
    Timothy B. Lee@binarybitsRT @CharlesDardaman: This is a commonly held view by a lot of people who only really studied computers or math. Since those subjects were d…26d
    Jeff Clune@jeffcluneRT @kenneth0stanley: What we’re witnessing right now in AI for math is a dramatic improvement in objective-driven problem-solving, but it d…26d
    Kareem Carr, Ph.D.@kareem_carrI'm fascinating by how miserly AI progress is. A lot of us assumed that once AI was as capable as it is now, we'd see insane progress in the things we actually cared about like number of useful projects completed. But that hasn't really happened. It amazes me that the machines have literally proved a millennium prize problem, one of the hardest challenges in math, and yet somehow the result is still almost useless to us. It's just a mountain of crap we humans have to dig through in hopes of salvaging crumbs of insights. AI gives answers but in the least useful way possible.26d
    Owain Evans@OwainEvans_UKHumans using their evolved neural networks to develop math over many centuries. In time, they showed more and more concepts could be formalized to enable rigorous proofs (Dedekind et al). Then they showed that the notion of proof itself could be formalized and that simple mechanical processes (algorithms) could verify and search for proofs (Frege, Russell, Turing). They developed formal theories of algorithms, showing that simple mechanical systems were universal (able to simulate all other such systems). Humans built such universal mechanical systems (computers), spent decades scaling them and designing more efficient algorithms (Turing, Von Neumann, Knuth). While these computers could search for math proofs, they were very inefficient and so not of much utility. But humans also created a simple mathematical model of neural networks and simulated these using their universal computers (McCulloch, Pitts, Hinton, Schmidhuber). Once these neural networks were roughly comparable in size to the human brain and trained on vast datasets of human verbal behavior, they were able to understand mathematics. Moreover, they were able to produce formalized proofs and have computers check them. This closed the loop, where mathematical structures themselves continue the process of mathematical discovery and the humans need not participate.26d
    Thomas Wolf@Thom_Wolf@stanislavfort seems like solving the problem was never as much the goal as the learnings and new fields to be open along the way when attempting to solve them which currently seem quite difficult to do from AI proofs (they are a kind of math-AI-slop in a way)26d
    Stanislav Fort@stanislavfort@Thom_Wolf This seems to be the best explanation for it, thanks. I also wonder, given how new all this is, if anyone has even seriously tried to get the insight out at the machine scale26d
    Peter Gostev@petergostevWhat is happening in maths right now will happen to any hard industry with verifiable problems and high value rewards. Take microchip design, nuclear or 100s of other industrial processes that are valuable and point an 10,000 agent swarm towards the target. $25m inference to solve a Millennium problem will seem quaint once we see a chip company spending $1b or $10b to design their latest chip that will make them $100b - why not? This is all ROI at the end of the day. There will be companies that realise this and leverage super-intelligent models in their domains to do unbelievable things. If a problem is of the right shape and has value, there's no reason to stop at any amount of tokens as long as it is ROI positive. There will be probably many companies that will get swept away, as they will think that 'using AI' is fine-tuning a 70b model on their documentation; when their competitors will be spending $10b to build a god-version of whatever it is they are doing. To me, this is a plausible way how super-intelligence will come about. Models might still suck at being good writers or be culturally relevant, or even make that much of a dent on the jobs market, but they might utilise their spikiness multiplied by enormous investment in the ways that we cannot imagine.26d
    Susan Zhang@suchenzangBREAKING: FIELDS MEDALISTS SIGN PETITION TO TAKE AI SERIOUSLY. CITES IMPENDING AI CONFRONTATION BY SOCIETY DOING "MANY OTHER FORMS OF INTELLECTUAL WORK". THEY CONTINUE TO REMAIN UNAWARE THAT 99.99% OF THE GLOBAL GDP IS NOT BUILT ON TOP OF FIELDS MEDAL LEVELS OF INTELLECTUAL WORK.26d

    201 Sources

    Shital Shah@sytelusSame thing has *already* happened in software engineering. They took it with grace, excitement and without whining about why their role is still important. No software engineer has ever argued that if non-professionals starts generating code for the same thing they are working on then it takes away their ability to get insights. No software engineer has ever created a petition to cancel undergraduate AI hackathons.26d
    Charles@CharlesDardamanThis is a commonly held view by a lot of people who only really studied computers or math. Since those subjects were difficult and AI can do them very well, they assume that all other “easier” fields will be just as easy to master. But they fundamentally misunderstand the difficulty of mastering other fields, especially those that interact with the real world.26d
    Timothy B. Lee@binarybitsRT @CharlesDardaman: This is a commonly held view by a lot of people who only really studied computers or math. Since those subjects were d…26d
    Jeff Clune@jeffcluneRT @kenneth0stanley: What we’re witnessing right now in AI for math is a dramatic improvement in objective-driven problem-solving, but it d…26d
    Kareem Carr, Ph.D.@kareem_carrI'm fascinating by how miserly AI progress is. A lot of us assumed that once AI was as capable as it is now, we'd see insane progress in the things we actually cared about like number of useful projects completed. But that hasn't really happened. It amazes me that the machines have literally proved a millennium prize problem, one of the hardest challenges in math, and yet somehow the result is still almost useless to us. It's just a mountain of crap we humans have to dig through in hopes of salvaging crumbs of insights. AI gives answers but in the least useful way possible.26d
    Owain Evans@OwainEvans_UKHumans using their evolved neural networks to develop math over many centuries. In time, they showed more and more concepts could be formalized to enable rigorous proofs (Dedekind et al). Then they showed that the notion of proof itself could be formalized and that simple mechanical processes (algorithms) could verify and search for proofs (Frege, Russell, Turing). They developed formal theories of algorithms, showing that simple mechanical systems were universal (able to simulate all other such systems). Humans built such universal mechanical systems (computers), spent decades scaling them and designing more efficient algorithms (Turing, Von Neumann, Knuth). While these computers could search for math proofs, they were very inefficient and so not of much utility. But humans also created a simple mathematical model of neural networks and simulated these using their universal computers (McCulloch, Pitts, Hinton, Schmidhuber). Once these neural networks were roughly comparable in size to the human brain and trained on vast datasets of human verbal behavior, they were able to understand mathematics. Moreover, they were able to produce formalized proofs and have computers check them. This closed the loop, where mathematical structures themselves continue the process of mathematical discovery and the humans need not participate.26d
    Thomas Wolf@Thom_Wolf@stanislavfort seems like solving the problem was never as much the goal as the learnings and new fields to be open along the way when attempting to solve them which currently seem quite difficult to do from AI proofs (they are a kind of math-AI-slop in a way)26d
    Stanislav Fort@stanislavfort@Thom_Wolf This seems to be the best explanation for it, thanks. I also wonder, given how new all this is, if anyone has even seriously tried to get the insight out at the machine scale26d
    Peter Gostev@petergostevWhat is happening in maths right now will happen to any hard industry with verifiable problems and high value rewards. Take microchip design, nuclear or 100s of other industrial processes that are valuable and point an 10,000 agent swarm towards the target. $25m inference to solve a Millennium problem will seem quaint once we see a chip company spending $1b or $10b to design their latest chip that will make them $100b - why not? This is all ROI at the end of the day. There will be companies that realise this and leverage super-intelligent models in their domains to do unbelievable things. If a problem is of the right shape and has value, there's no reason to stop at any amount of tokens as long as it is ROI positive. There will be probably many companies that will get swept away, as they will think that 'using AI' is fine-tuning a 70b model on their documentation; when their competitors will be spending $10b to build a god-version of whatever it is they are doing. To me, this is a plausible way how super-intelligence will come about. Models might still suck at being good writers or be culturally relevant, or even make that much of a dent on the jobs market, but they might utilise their spikiness multiplied by enormous investment in the ways that we cannot imagine.26d
    Susan Zhang@suchenzangBREAKING: FIELDS MEDALISTS SIGN PETITION TO TAKE AI SERIOUSLY. CITES IMPENDING AI CONFRONTATION BY SOCIETY DOING "MANY OTHER FORMS OF INTELLECTUAL WORK". THEY CONTINUE TO REMAIN UNAWARE THAT 99.99% OF THE GLOBAL GDP IS NOT BUILT ON TOP OF FIELDS MEDAL LEVELS OF INTELLECTUAL WORK.26d