Earlier this month, WIRED published an excerpt from Steve Rosenbaum’s trendy fresh book, The future of truthwhich looks at how artificial intelligence distorts people’s sense of reality. The Fresh York Times soon followed. reported that the book contained more than half a dozen invented or misattributed quotes. In a statement, Rosenbaum, who earned a master’s degree in “truth” from Fresh York University, admitted that he accidentally included “a handful” of “misattributed or synthetic” quotes. Ironically, the veracity of a book about the impact of artificial intelligence on truth has now come under intense scrutiny because of its author’s exploit of artificial intelligence.
After the Los Angeles story came to delicate, WIRED took another look at our 1,450-word piece. The fact-checking team reviewed it before publication and we have re-confirmed that the quotes and facts it contains are true. However, WIRED’s editorial policy on generative artificial intelligence prohibits the publication of texts generated and edited by artificial intelligence, and a reader email stating that a piece was “blatantly written by artificial intelligence” raised further questions about the extent to which Rosenbaum used artificial intelligence tools. IN The future of truthin the acknowledgments section, Rosenbaum writes that ChatGPT, Claude, NaturalReaders, ProWritingAid, and Grammarly helped “refine and refine the presentation [his] ideas.” What exactly did that mean?
WIRED ran the snippet through several artificial intelligence detection services, including Pangram, GPTZero and ZeroGPT. Each service suggested that it was probably AI-generated or was AI-generated with high confidence. However, AI detection tools are unreliable and may return inaccurate readings. So WIRED’s research director emailed Rosenbaum directly, asking if and how he used artificial intelligence to write the piece.
He wrote back: “Like many writers working today, I used AI tools during parts of the book’s research and editing process, including source discovery, brainstorming, structured feedback, and language refinement.” However, he stressed that “the ideas, reports, arguments and final authorship are mine and the WIRED piece was not generated by artificial intelligence and then simply published as is.” He urged WIRED editors to be cautious and trust artificial intelligence detection tools, noting that false positives can occur.
At this point, WIRED’s senior editors asked me to look at this episode because I’ve been covering various forms of AI bugs since 2024. My first step was to check the entire text of the book using the Pangram detection tool. (While all AI detection tools have limitations and may produce false positives, the current gold standard is Pangram). It found that 53% of the book was AI-generated, with an additional 9% recorded as possibly AI-powered.
I called Rosenbaum and asked for a more detailed description of how he used artificial intelligence to write the book and whether he disputed Pangram’s results. (BenBella Books, whose imprint published The future of truthdid not respond to requests for comment. Simon & Schuster, the company that distributes BenBella’s books in the United States, declined to comment.)
Rosenbaum would not comment on the accuracy of Pangram’s results. In fact, he didn’t want to talk about them at all. “I am not participating in this conversation,” he said. “It’s like saying: do you beat your wife? This is one of those accusations that has no answer.”
Instead, he offered a broad explanation of his editorial process. He says he used artificial intelligence tools as search engines early in the writing process, which helped him find information for more research-intensive sections of the book. To demonstrate how he would do this, he asked ChatGPT to describe me and then read the results aloud. AI Search has more or less accurately described some of my previous stories, including work on AI-generated “zombie media sites.”