From the even-a-stopped-clock-is-right-twice-a-day department:
FTC Chair Andrew Ferguson told Reuters that he thinks that companies should be liable for chatbot harms. At that time, Ferguson emphasized that he was tired of people anthropomorphizing AI tools. He insisted that, “if someone tells a tool to do something, and the tool does it, I don’t think we would say, ‘Oh, what do we do about the tool?’”
“But we didnt mean for those things to happen.” “We didnt design it for that.” “We cant be held responsible for the AI going rogue.”
Bro, if I sold you a chat bot that sometimes committed phishing attacks and stole your identity, do you think “We just wanted it to make it really good at personalized conversations by getting to know you” frees me from the liability of how it determined it should do that and what it was capable of doing to follow through with it?
This shit is happening in your own labs. You can’t isolate and maintain control over systems that are 100% under your ownership and management with a team of tech experts, and you think releasing this stuff to the general public won’t result in 1000 times worse outcomes? How is that not criminal negligence?