Scientists Train AI to Be Evil, Find They Can't Reverse It

Bad AIpple How hard would it be to train an AI model to be secretly evil? As it turns out, according to some researchers, not very — and attempting to reroute a bad apple AI's more sinister proclivities might backfire in the long run. In a new paper published to the pre-print server ArXiv, researchers […]

ai ai model ai-safety anthropic apple apple ai artificial intelligence paper researchers scientists the digest train train ai