Interestingly, I think if we were doing a better job of identifying abusive and asshole comments on our giant social media platforms, we could build a kind of i

Interestingly, I think if we were doing a better job of identifying abusive and asshole comments on our giant social media platforms, we could build a kind of internal filter for such a chat bot. Learn an "asshole comment" classifier, and then have the chat bot ask itself, will I sound like an asshole before each utterance... As a bonus, the classifier would also be great for providing social media moderation tools, at the same time.