The vast majority of her comrades at HuffPost or any such rag headquartered in Manhattan would not be so adventurous — in fact, as she notes, they warned her in the sternest terms that she was risking her life agreeing to meet a domestic terrorist for lunch at a busy public diner.

The journey begins with this woman, Jen Golbeck — "a professor at the University of Maryland where she studies extremism, social media, malicious online behavior, and artificial intelligence" —who has apparently been damaged by spending way too much time online, refusing to hold out the possibility that a very real person she is interacting online with is, in fact, a real person.

Via HuffPost (emphasis added):

"I was pretty sure this tweet, sent in response to a recent essay I published on HuffPost detailing my summer interviewing people at Donald Trump rallies, was written by a bot. It had all the telltale signs: mudslinging; a username consisting of a first name followed by a bunch of numbers; and the "MAGA" sign-off.

"Thank you for your comment, name bunchanumbers!" I replied ? to which the account responded: "You are welcome. I have lived in MD a long time and have never seen anyone as biased and full of shit as you are. MAGA."

We continued to tweet back and forth at each other a few times. I asked for croissant recipes and answers to math questions in hopes of getting the AI I believed to be running the bot to reveal itself. Instead, I eventually got a reply I never saw coming: "Let me take you to lunch next week and discuss how you can help Make America Great Again! My treat. MAGA."

This was a first for me. As a journalist and researcher who has spent much of the past decade studying the psychology of far-right movements, I'd interacted with countless MAGA die-hards, but none of them had ever invited me to a meal. I was intrigued, so I agreed, and within an hour we had exchanged cell numbers and made a date."