AI300
  1. 211.
    @GoodFaithOnly @binarybits When I say AGI I mean a "feature complete" build of a remote worker with a lot of 9s (think: a bit like current Autopilot FSD, maybe plus a few more iterations).›
    Andrej Karpathy (AI researcher) · LLM score 70 · 3 months ago
  2. 212.
    Second @simonbatzner.›
    Xiang Fu (Researcher at Periodic Labs) · LLM score 80 · 3 months ago
  3. 213.
    @lineardiff @ericzelikman I don’t view that as incompatible.›
    Noam Brown (OpenAI Research Scientist) · LLM score 20 · 3 months ago
  4. 214.
    new paper "ATLAS" – for details: https://t.co/MJO9mvlS90; it also covers a new scaling law, pretrain vs finetune for multilingual & more :)›
    Niklas Muennighoff (AI Researcher at Stanford) · LLM score 70 · 3 months ago
  5. 215.
    To scale data-constrained LLMs, repeating & denoising objectives can help.›
    Niklas Muennighoff (AI Researcher at Stanford) · LLM score 85 · 3 months ago
  6. 216.
    State space architecture for state of the art voice model! https://t.co/QjdOdrIFSM
    Tri Dao (Chief Scientist at Together) · LLM score 80 · 3 months ago
  7. 217.
    To adapt to today's world, crank down discount factor and crank up learning rate
    Xiang Fu (Researcher at Periodic Labs) · LLM score 75 · 3 months ago
  8. 218.
    Happy to share a new paper! Designing model behavior is hard -- desirable values often pull in opposite directions.›
    John Schulman · LLM score 80 · 3 months ago
  9. 219.
    Every AI researcher should read two essays.›
    Xiang Fu (Researcher at Periodic Labs) · LLM score 80 · 3 months ago
  10. 220.
    Below is a deep dive into why self play works for two-player zero-sum (2p0s) games like Go/Poker/Starcraft but is so much harder to use in "real world" domains.›
    Noam Brown (OpenAI Research Scientist) · LLM score 80 · 3 months ago
  11. 221.
    @pli_cachete @karpathy Because it means self play converges to a minimax equilibrium which is essentially an unbeatable strategy.›
    Noam Brown (OpenAI Research Scientist) · LLM score 80 · 3 months ago
  12. 222.
    Self play works so well in chess, go, and poker because those games are two-player zero-sum.›
    Noam Brown (OpenAI Research Scientist) · LLM score 75 · 3 months ago
  13. 223.
    .@Stanford courses are high-quality but the policies are definitely outdated.›
    Noam Brown (OpenAI Research Scientist) · LLM score 20 · 3 months ago
  14. 224.
    Awesome to see Veo 3.1 top the LMArena video leaderboards by a large distance with big improvements over Veo 3.0 for text-to-video (+30) and image-to-video (+70)! 🔥Huge congrats to the team! Try it for yourself in https://t.co/QgTpxTL8DQ and the @GeminiApp https://t.co/nQ14ZnkKnr
    Demis Hassabis (CEO of DeepMind) · LLM score 20 · 3 months ago
  15. 225.
    @fchollet Didn't you say just two months ago that you think AGI is about 5 years away?›
    Noam Brown (OpenAI Research Scientist) · LLM score 20 · 3 months ago
  16. 226.
    Super excited to be collaborating with Commonwealth Fusion Systems @CFS_energy to use AI to accelerate fusion development - and move closer to a sustainable future with limitless clean energy https://t.co/kCS1UKEtjP
    Demis Hassabis (CEO of DeepMind) · LLM score 70 · 4 months ago
  17. 227.
    Veo 3 is the state-of-the-art in video models.›
    Demis Hassabis (CEO of DeepMind) · LLM score 30 · 4 months ago
  18. 228.
    We processed over 1.3 Quadrillion tokens last month - that's 1,300,000,000,000,000 tokens! or to put it another way that's 500M tokens a second or 1.8 Trillion tokens an hour...›
    Demis Hassabis (CEO of DeepMind) · LLM score 75 · 4 months ago
  19. 229.
    Some generous companies in Toronto are funding three lectures on AI safety by Owain Evans on Nov 10, 11, 12.›
    Geoffrey Hinton · LLM score 60 · 4 months ago
  20. 230.
    This work, led by @_junxiong_wang and @ben_athi, is a first step towards building AI systems that evolve and get better as you use them.›
    Tri Dao (Chief Scientist at Together) · LLM score 20 · 4 months ago
  21. 231.
    @itsandrewgao Worst-of-N is a good solution to this.›
    Noam Brown (OpenAI Research Scientist) · LLM score 85 · 4 months ago
  22. 232.
    More info on some the cool capabilities of Genie 3 here: https://t.co/wcKztrUSu6
    Demis Hassabis (CEO of DeepMind) · LLM score 40 · 4 months ago
  23. 233.
    Fantastic to see Genie 3, our state-of-the-art world model, featured in @TIME's 2025 Best Inventions.›
    Demis Hassabis (CEO of DeepMind) · LLM score 40 · 4 months ago
  24. 234.
    @littmath I decided to check with a doctor friend and they told me “NSAIDs have a tendency to increase your risk of clotting (despite the fact that they affect the functionality of your platelets resulting in a tendency to bleed)”
    Noam Brown (OpenAI Research Scientist) · LLM score 40 · 4 months ago
  25. 235.
    @littmath Can you point out what specifically in the GPT-5 Thinking response you find problematic? Because so far you’ve not provided any evidence that its critique of “prevents blood clots” is unjustified
    Noam Brown (OpenAI Research Scientist) · LLM score 20 · 4 months ago
  26. 236.
    @littmath I'm confused why you're so confidently taking a strong position on this despite admitting that you're not a doctor.›
    Noam Brown (OpenAI Research Scientist) · LLM score 20 · 4 months ago
  27. 237.
    Are these just hallucinations by GPT-5 Thinking? Nope.›
    Noam Brown (OpenAI Research Scientist) · LLM score 40 · 4 months ago
  28. 238.
    "One of our goals is to discover superconductors that work at higher temperatures than today's materials"›
    Noam Brown (OpenAI Research Scientist) · LLM score 70 · 4 months ago
  29. 239.
    Bullish, in the coming decades majority of compute will be spent on ai for science https://t.co/8aLDhazs6i
    Jason Wei (AI Researcher at Meta) · LLM score 70 · 4 months ago
  30. 240.
    @j2bryson I know some AI startups that I think are way overpriced.›
    Noam Brown (OpenAI Research Scientist) · LLM score 30 · 4 months ago
More