Research acceleration: The view inside OpenAI

(openai.com)

31 points | by iamsyr 2 hours ago

3 comments

  • Jeff_Brown 46 minutes 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.
    • grim_io 26 minutes ago
      They would maybe try to deactivate that bad "gene" and move on, exposing future models to "genetic disorders".
    • coherentpony 22 minutes ago
      “All models are wrong. Some are useful.” - George Box
  • simonw 48 minutes 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 0 minutes ago
      Indeed, many programmers might pattern match to repetitive stress injury and think of their brushes with carpal tunnel syndrome. :)
    • vatsachak 10 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 4 minutes ago
        Are you sure that was not iterative improvement?
  • matan0904 21 minutes ago
    [flagged]