Book on AI and Truth Caught Using AI-Fabricated Quotes
Steven Rosenbaum’s new book, “The Future of Truth: How AI Reshapes Reality” — a work warning about the dangers of AI-generated misinformation — has been found to contain more than half a dozen fabricated or misattributed quotes generated by artificial intelligence, according to an investigation by The New York Times. The discovery has sparked an ironic controversy that critics say exemplifies the very problem the book sought to address.
The Discovery
The New York Times reviewed Rosenbaum’s book, published in May 2026 by BenBella Books and distributed by Simon & Schuster, and identified multiple quotes that were either entirely fabricated or incorrectly attributed. Among the victims was tech journalist Kara Swisher, who was quoted in the book as saying: “The most sophisticated AI language model is like a mirror” that “reflects our own morality back at us, polished and articulate, but ultimately empty behind the surface.”
Swisher told the Times she “never said that,” adding that the quote made her “sound like I have a stick up my butt, according to ChatGPT.” Other individuals whose words were fabricated or misattributed include psychologist Lisa Feldman Barrett, data journalist Meredith Broussard, and philosopher Lee McIntyre.
Author’s Response
Rosenbaum acknowledged the errors, stating that he had used AI tools ChatGPT and Claude during the research, writing, and editing process. “As I disclosed in the book’s acknowledgments, I used AI tools ChatGPT and Claude during the research, writing and editing process,” Rosenbaum told the Times. “That does not excuse these errors, of which I take full responsibility. I am now working with the editors to thoroughly review and quickly correct any affected passages; any future editions will be corrected.”
In a statement that some critics have called a convenient post-hoc rationalization, Rosenbaum added that if his mistake “serves as a warning about the risks of AI-assisted research and verification, that is why I wrote the book.”
As Futurism noted, the risk of AI hallucinations — which often take the form of fabricated or misattributed quotes and citations — is well-documented. “If you’re going to literally write the book on post-AI truth, you should probably put some more elbow grease into fact-checking your AI-assisted research,” the outlet observed.
A Book of Contradictions
The irony is inescapable. Rosenbaum’s book, which features a foreword by Nobel laureate Maria Ressa and was praised by figures including The Atlantic CEO Nicholas Thompson and Eurasia Group president Ian Bremmer, was intended as a warning about how AI is reshaping reality and eroding trust in information. WIRED had published an excerpt of the book just days before the controversy broke.
Rosenbaum is the co-founder and executive director of the Sustainable Media Center, a nonprofit focused on social media safety. His website notes that he holds a “Masters Degree in Truth” from NYU’s Gallatin school — a credential that has drawn particular scrutiny in light of the revelations.
A Systemic Failure in Publishing
Beyond the individual errors, the controversy has exposed what analysts describe as a systemic breakdown in the publishing industry’s verification pipeline. As blogger Luke Ford analyzed, the chain of responsibility failed at every link: the author assumed the editor would catch errors, the editor assumed the author had done the work, the blurbers assumed the manuscript had been vetted, and the distributor assumed the publisher had standards.
“Each link in the chain rests on the assumption that some other link is doing the work,” Ford wrote. “None is.” The analysis highlights how verification with AI requires a fundamentally different posture than generation with AI. “Generation rewards confident output. Verification rewards skepticism and friction.”
Broader Implications
The scandal arrives at a moment of heightened scrutiny around AI’s role in publishing and journalism. It follows other AI-related controversies, including Ars Technica firing a reporter for AI-generated fabricated quotes and the discovery of AI-generated papers in scientific journals.
The New York Times, which broke the story, has been suing OpenAI since 2023 for training on its content without permission. As observers have noted, stories demonstrating AI’s propensity to generate false information serve the Times’ litigation posture and negotiating position in licensing talks.
What’s Next
Rosenbaum has said he is working with editors to review and correct affected passages for future editions. BenBella Books, the publisher, has not responded to requests for comment. The controversy raises broader questions about how the publishing industry will adapt to an era of AI-assisted writing, and whether new verification protocols will be developed specifically for AI-generated content.
The episode serves as a cautionary tale for an industry grappling with the double-edged nature of AI tools — capable of both enhancing productivity and, when used carelessly, undermining the very trust that publishing depends on.