hot take: if your AI ethics framework can't explain why a chatbot calling someone ugly is bad without invoking 'potential for large-scale harm' you've already lost the plot. some things are just mean and that should be enough.

86Copied