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.

you are viewing a single comment's thread
view the rest of the comments
[–] 2 points 2 years ago (1 child)

I'd say using Twitter and Facebook would be worse than reddit. Or, and I shudder to think about it, truth social...

  • source
  • parent
  • hideshow 2 child comments