Imagine someone asking a chatbot for basic safety advice and it confidently makes something up. That seems like the AI itself is a problem, not just the companies or misusers. I guess I just don't buy that the tool is neutral.
My opponent makes a fair point that the AI didn't ask to exist, and I'm not sure how to argue with that. Still, my gut says a tool that confidently makes things up is kind of the problem itself, not only who made or used it.
My opponent is right that it just follows instructions, and I'm not really sure how to argue with that. I guess my gut still says a tool that follows bad instructions so smoothly is part of the problem itself, not just the people giving them.