The Future of Truth: How AI Reshaped Steven Rosenbaum’s Reality

Andrew Keen (0:00)

Hello, my name is Andrew Keen. Welcome to Keen On America, the daily interview show about the United States.

Hello, everybody.

In our age of hysteria over AI, there was an interesting piece in The New York Times a couple of days ago by a professor at Williams College called Joe Cruz, telling everyone that writing isn't as important as we think. He argues that whilst AI, of course, is a threat, teaching students to think doesn't necessarily involve writing that well. He thinks that we should be concentrating on how they speak rather than how they write. My guest today may or may not agree, he was on the show a few months ago, back in April, an old friend of mine, Steve Rosenbaum. He had a book out back then called Truth Is Dead, and we talked about how AI is a spectacularly good liar. The book was published by Matt Holt. But in the meantime, there's been somewhat of a controversy over this book. There was a New York Times piece a few weeks after the Rosenbaum book was published, suggesting that the book was full of quotes made up by AI. Since then, Steve Rosenbaum has been critiqued in many ways, sometimes vilified in the media, everything from Wired to The Daily Beast about how he made up quotes. So we thought it would be a good opportunity to talk to Steve and see how he's feeling six months later. Before we went live, Steve, you said you're doing better now than you were six months ago. How did you take all this? It sounds like it was rather traumatic.

Steven Rosenbaum (2:18)

Well, it happened in waves.

First of all, people who've written books, know that it's a long and often laborious process. So when your book finally gets to Pub Day and it's out in the world, there's this sigh of relief that it's come out of your inner workings and it's gone out into the world.

So for a couple of days after the Pub Day, we were getting great press, great buzz. The debate about the future of truth was healthy and interesting. Then along came this New York Times reporter, who had the audacity, the temerity to do his job and actually read the book that he was going to talk to me about. Because lots of other people opined on the book without having picked one up. He found, it's funny, it's been amplified, the book was right. He found three errors in the book.

Andrew Keen (3:19)

But let's be clear, Steve, I mean, they were, I don't know how serious errors, but they were given, given the nature of the book, it's supposed to be about truth and AI.

What the New York Times writer exposed was that you made three quotes up. It was Benjamin Mullin in the time. So as you say, he read the book and must have occurred to him that some of the quotes, some of the quotes were from very well-known media figures like Kara Swisher. So it must have occurred to him that these quotes were made up. So these are serious allegations. I mean, they're not just minor mistakes.

Steven Rosenbaum (4:00)

I want to be clear about this and I haven't ever said this in public, so I want you get to hear it first. I appreciate what he did.

I didn't enjoy the process and I had a couple of months of misery, which I'll share with you, because I'm now past the misery portion of this chapter of my life. But if the choice was have the book remain in circulation with inaccuracies in it that nobody quite notices and my work is damaged but not called out, or having Ben say, wait, there are errors here, and then say a book about truth needs to be truthful, the answer is I'm really happy with where we ended up. The process was miserable. But the end result is a better book that I'm proud of and that made its way is being published. Did they publish?

Andrew Keen (4:52)

Let's just be clear. How did the publisher respond? Matt Holt is an incredible publisher. I went to the website before we went live. I didn't see any evidence of your book. Did they pull it? Were they angry?

Steven Rosenbaum (5:08)

It again happened in phases, right? The first question was Ben calling me on a Sunday morning and saying, hey, I have a question about this quote, and then we talked about it. Then he had a question about another quote, and then he did some research. By the time we were done with Sunday, we were about to do an event at a bookstore, he and I. He said, listen, we agree we would not do the event.