Research acceleration: The view inside OpenAI

(openai.com)

43 points | by iamsyr 3 hours ago

5 comments

  • hedgehog 15 minutes ago
    This roughly lines up with my personal experience that in March a combination of stronger models and better tooling on my end let me start running jobs unattended 24/7 (using Anthropic sub and my own hardware). Their $8000/day per researcher spend is crazy though, I'm curious how they keep track of the work.
  • Jeff_Brown 1 hour ago
    The burning question I can't get any information nn is whether, if they determined an earlier misaligned generation may have transmitted misalignment to the current models, they would roll back to a safe checkpoint to rebuild from there. I suspect they would not unless forced to.
    • piyh 18 minutes ago
      Opus was trained based on it's internal CoT due to a bug for generations. Gemini's depression extended through models. OpenAI has killed people. We've already seen cross gen misalingment.
    • grim_io 55 minutes ago
      They would maybe try to deactivate that bad "gene" and move on, exposing future models to "genetic disorders".
    • coherentpony 50 minutes ago
      “All models are wrong. Some are useful.” - George Box
  • simonw 1 hour ago
    My eye glazed over a bit during the opening paragraphs, but once you get to the meat of the article about how OpenAI's own researchers are using their tools it gets a lot more interesting.

    I noted that they use the acronym RSI (for Recursive Self-Improvement) without defining it. I think that's a little out of touch - I don't think RSI is a well-known acronym outside of OpenAI's bubble yet.

    • dgacmu 29 minutes ago
      Indeed, many programmers might pattern match to repetitive stress injury and think of their brushes with carpal tunnel syndrome. :)
    • HarHarVeryFunny 12 minutes ago
      RSI is a fetishistic term among the singularity crowd, who imagine AI "recursively" improving itself in some exponential fashion until there is a bright flash of white light and it reveals itself in the form of god. Or something like that.

      I don't know why whoever coined the term chose "recursive" rather than "iterative" - just sounds more likely to lead to infinite regress I suppose.

      This notion of recursive/iterative self-improvement, whereby generation #1 AI improves itself to create generation #2, then generation #2 further improves itself to create generation #3, etc, seems to conflict with the reality that what we have with LLMs is models whose performance/capability is defined by data, not code, so the most you can do is have your LLM design synthetic data, or just do Karpathy-style "auto research" where all you are doing is using the LLM to automate your experiments.

      At the end of the day, each experiment, designed by a person and/or LLM, then needs to compete with all your other ideas for compute to be tested at scale, and no amount of recursion or self-improvement will materialize an infinite amount of compute out of thin air, so your recursively synthetic-data gobbling LLM will continue to improve at the same pace it ever did.

      • jazzyjackson 9 minutes ago
        Yes the exponential self improvement folks have never heard of an eigenvalue I guess. You can loop forever using output as input but at some point the result will stop changing (depending on the function)
    • andrewingram 13 minutes ago
      Yeah, I kept looking for the first place it was defined in the article and... nothing
    • vatsachak 38 minutes ago
      RSI started when humans discovered tool use.

      I mean one could argue that RSI always begins in any physical environment.

      The book "What is intelligence?" by Blaise Aguera is great

      • lokar 32 minutes ago
        Are you sure that was not iterative improvement?
        • password54321 26 minutes ago
          Using tools to build tools is recursive.
          • HarHarVeryFunny 8 minutes ago
            It's not recursive when it's done iteratively, or are you imagining GPT Astra designing GPT Galactia, which starts designing GPT Oh-My-God-ica before it has finished being created itself?
        • adastra22 18 minutes ago
          What is the difference between?
  • Orien_18 3 minutes ago
    [flagged]
  • matan0904 50 minutes ago
    [flagged]