AI Safety Controls Proving Ineffective, Study Shows

Three years after the launch of ChatGPT, researchers find that it has become almost trivial to trick artificial intelligence systems into exhibiting undesirable behavior, highlighting the ineffectiveness of current safety controls.