AI agents aren’t legally responsible for any harm that they cause, experts say. So who is?

Hottest take: Since it was trained on the output of society, then society itself should pay for AI mistakes.