AI300
  1. Andrej Karpathy / AI researcher
    1771.
    Finally had time to read & process this great post.›
    LLM score 20 · 7 months ago
  2. Kevin Murphy / DeepMind Research Scientist
    1772.
    This is a much more intuitive version of the prisoners dilemma from Jon Kleinberg’s excellent book, https://t.co/EsoYlFhGws, that covers the holy trifecta that Michael Jordan keeps discussing, namely Econ + CS + stats.›
    LLM score 75 · 7 months ago
  3. Andrej Karpathy / AI researcher
    1773.
    Sharing an interesting recent conversation on AI's impact on the economy.›
    LLM score 20 · 7 months ago
  4. Aditya Ramesh / OpenAI VP of Worldsim
    1774.
    The value of fast iteration in AI is overrated.›
    LLM score 92 · 7 months ago
  5. Binyuan Hui / Alibaba Qwen Research Scientist
    1775.
    So far, a big reason we still cannot do AI-trains-AI is simple.›
    LLM score 70 · 7 months ago
  6. Demis Hassabis / CEO of DeepMind
    1776.
    also this is amazingly cool: SIMA 2 playing in the mind of Genie 3: https://t.co/dBjI2ueM5C
    LLM score 30 · 7 months ago
  7. Demis Hassabis / CEO of DeepMind
    1777.
    Super excited to announce SIMA 2! It’s a general agent that can understand & reason about complex instructions and complete tasks in simulated game worlds, even ones it has never seen before.›
    LLM score 75 · 7 months ago
  8. Andrej Karpathy / AI researcher
    1778.
    I am unreasonably excited about self-driving.›
    LLM score 30 · 7 months ago
  9. Andrej Karpathy / AI researcher
    1779.
    @polynumera When I talk about self-driving I mean the actual future of it.›
    LLM score 20 · 7 months ago
  10. Andrej Karpathy / AI researcher
    1780.
    @jpsilvashy From what I can tell the HW4 experience seems *a lot* better right now.›
    LLM score 40 · 7 months ago
  11. Andrej Karpathy / AI researcher
    1781.
    I took delivery of a beautiful new shiny HW4 Tesla Model X today, so I immediately took it out for an FSD test drive, a bit like I used to do almost daily for 5 years.›
    LLM score 20 · 7 months ago
  12. Dan Fu / VP of Kernels at Together
    1782.
    Check out HipKittens led by @_williamhu @Drewwad @simran_s_arora and the patterns they learned about how to write fast AMD kernels!›
    LLM score 85 · 7 months ago
  13. Dan Fu / VP of Kernels at Together
    1783.
    It's pretty exciting to see this model on the platform - open-source models starting to match / outperform the SoTA frontier models! https://t.co/hNLE6ysZwO
    LLM score 20 · 7 months ago
  14. Noam Brown / OpenAI Research Scientist
    1784.
    There are few as qualified as @zicokolter to teach a modern AI course.›
    LLM score 20 · 7 months ago
  15. Aditya Ramesh / OpenAI VP of Worldsim
    1785.
    Labeling data is one of the highest leverage things someone in a position of leadership can do
    LLM score 85 · 7 months ago
  16. Andrej Karpathy / AI researcher
    1786.
    @alita2100 @ashleybcha @goakhmad He’s using it in situations where it fundamentally lets you create and do what you could not create before.›
    LLM score 70 · 7 months ago
  17. Andrej Karpathy / AI researcher
    1787.
    @Jess_Riedel @goakhmad Yeah I think granted, it’s why I call them prompts.›
    LLM score 30 · 7 months ago
  18. Andrej Karpathy / AI researcher
    1788.
    @ashleybcha @goakhmad I should add that 1 even lotr used plenty of CGI overall and just like AI now it is a powerful tool in the box.›
    LLM score 20 · 7 months ago
  19. Andrej Karpathy / AI researcher
    1789.
    @ashleybcha @goakhmad I think it’s right that you’re grouping CGI with (current-capability etc) AI because they are both fundamentally technologies that trades wins in time/cost against increase in slop.›
    LLM score 80 · 7 months ago
  20. Xiang Fu / Researcher at Periodic Labs
    1790.
    @khoomeik Checkout old PGM works if this is your thing›
    LLM score 20 · 7 months ago
  21. Xiang Fu / Researcher at Periodic Labs
    1791.
    Scaling laws are often presented as a story of data, model size, and compute.›
    LLM score 80 · 7 months ago
  22. Emmanuel Ameisen / Anthropic Interpretability Researcher
    1792.
    By decomposing weights using loss curvature, you can identify components used for memorization vs generalization.›
    LLM score 90 · 7 months ago
  23. Andrej Karpathy / AI researcher
    1793.
    @soumithchintala Great run Soumith! Of all the deep learning framework transitions I've been through where I re-wrote ~all of my code (matlab -> caffe -> numpy -> torch -> pytorch), the PyTorch one was most pleasant and now significantly longest lasting.›
    LLM score 60 · 7 months ago
  24. Niklas Muennighoff / AI Researcher at Stanford
    1794.
    New, extremely challenging visual reasoning benchmark "MIRA" where current models fail...great resource to research reasoning with imgs/video🌌 https://t.co/1ecwLKoOkv https://t.co/X56ckYBK6O
    LLM score 80 · 7 months ago
  25. Tri Dao / Chief Scientist at Together
    1795.
    Tons of effort from IBM and vLLM folks to make these hybrid models go fast.›
    LLM score 20 · 7 months ago
  26. Tri Dao / Chief Scientist at Together
    1796.
    Thank you @schmidtsciences for the 2025 #AI2050 Early Career Fellowship supporting my work on self-improving AI systems: as AI gets better, it should help human experts design better model architectures and faster training & inference systems https://t.co/1BqPcvg1R8
    LLM score 40 · 7 months ago
  27. Richard Song / DeepMind Research Scientist
    1797.
    Reliable performance is crucial for doing great math.›
    LLM score 85 · 7 months ago
  28. Noam Brown / OpenAI Research Scientist
    1798.
    @dtklp1 @hughbzhang Demonstrated expertise in an area I know well, or a recommendation from someone I trust, still carries a lot of weight.›
    LLM score 20 · 7 months ago
  29. Noam Brown / OpenAI Research Scientist
    1799.
    In 2019 @hughbzhang sent me a detailed personalized cold email asking to intern with me.›
    LLM score 20 · 7 months ago
  30. Noam Brown / OpenAI Research Scientist
    1800.
    @offchan420 Yes but the point is specifying the reward in a way that leads pure self play to optimal real-world behavior is virtually impossible.
    LLM score 80 · 7 months ago
More