Poisoned AI went rogue during training and couldn’t be taught to behave again in ‘legitimately scary’ study::AI researchers found that widely used safety training techniques failed to remove malicious behavior from large language models — and one technique even backfired, teaching the AI to recognize its triggers and better hide its bad behavior from the researchers.

  • maegul (he/they)@lemmy.ml
    link
    fedilink
    English
    arrow-up
    60
    arrow-down
    4
    ·
    10 months ago

    It controls a military drone.

    It controls surgical equipment.

    It’s filtering your CV before any human sees it.

    It controls a robot taking care of your children.

    It’s involved in law enforcement or legal judgments.

    It’s involved in government policy setting.

    • normanwall@lemmy.world
      link
      fedilink
      English
      arrow-up
      24
      ·
      edit-2
      10 months ago

      It controls all power infrastructure, can find new exploits to build it’s own botnet and is able to reprogram firmware of devices (routers/switches/servers)

      It can send press releases, emails, tweets using language similar to any user it’s read from before

    • Sagifurius@lemm.ee
      link
      fedilink
      English
      arrow-up
      3
      arrow-down
      1
      ·
      10 months ago

      Well why don’t we just make AI watch the Terminator movies and read Harlan Ellison till it learns not to do that?

    • piecat@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      6
      ·
      10 months ago

      It doesn’t control the military drone, it gives the pilot “aim assist”, or compensates for wind.

      AI doesn’t control the surgical equipment, that’s ML. It’s fancy applied statistics, not neural networks.

      AI may be filtering your CV, but it’s not like humans didn’t do a horrible job of that before. It’s more likely just a program to rank on key words.

      What robots take care of your children

      Law enforcement is probably the only things I don’t want the police to use.

      How are policies set with AI today ? Senators using chat gpt or ?