He misrepresented himself as a big fan of all these blogs, who's read their posts etc. and that's how he achieved such a high response rate. In effect he deceived people into trusting him enough to spend their time on a response.
Now ordinarily this would be a little "white lie" and probably not a huge deal, but when you multiply it by telling it 1,000 times it becomes a more serious issue.
This is already an issue in email marketing. The gold standard of course is emailing people who are double opted in and only telling the truth, and if AI is used to help create that sort of email I don't really have a problem. There is basically a spectrum where the farther away you get from that the progressively more illegal/immoral your campaigns become. By the time you are shooting lies into thousands of inboxes for commercial purposes... you are the bad guy.
Sorry to say but the real issue here is Kurt has crossed an ethical line in promoting his startup. He did the wrong thing and he could have done it pretty effectively with conventional email tools too.