The Irony of The Future of Truth: AI Fake Quotes and the Truth About Gen Z

Steven Rosenbaum, the author of a nonfiction book exploring the impact of artificial intelligence on the concept of truth, has acknowledged that his work contains numerous fake or misattributed quotes generated by AI.

The book, titled The Future of Truth, was released in May 2026. Following a review by The New York Times, Rosenbaum admitted on May 18, 2026, that he had incorporated more than a half-dozen fabricated quotes in sections of the text.

The discovery highlights a critical failure in the editorial and research process, specifically regarding the reliance on large language models (LLMs) for primary verification and factual sourcing.

AI-Generated Fabrications in Nonfiction

The New York Times reported on May 19, 2026, that the misattributed or entirely concocted quotes were the result of AI tools used during the writing process. The errors appeared in multiple sections of the book, which had been released with significant promotion.

The incident underscores the known tendency of LLMs to hallucinate—a technical phenomenon where AI generates plausible-sounding but entirely false information—when tasked with retrieving specific citations or quotes.

In response to the findings, Rosenbaum attempted to frame the editorial errors as an illustration of the book’s core thesis.

In his statement, Mr. Rosenbaum said that if the episode serves as a warning about the risks of A.I.-assisted research and verification, that is why I wrote the book.The New York Times

These A.I. Errors do not, in fact, diminish the larger questions that the book raises about truth, trust and A.I. And its impact on society, democracy and editorial, he added.The New York Times

Contradictions in Epistemic Analysis

The controversy is compounded by Rosenbaum’s previous commentary on how different generations perceive truth. In an excerpt published in Wired, Rosenbaum argued that Generation Z has developed different epistemic habits, moving away from traditional institutional gatekeepers in favor of communal verification through social networks and emotional cues.

While the Wired piece presented a nuanced view of how digital natives navigate information, the subsequent revelation of AI-fabricated quotes in his own book suggests a failure to apply those same rigorous verification standards to his own research.

The use of AI to generate primary content without human oversight represents the most high-risk application of LLMs in professional writing. While these tools can assist in structuring or refining text, using them as a substitute for primary research often leads to the dissemination of false information.

The Risks of AI-Assisted Research

This case serves as a technical cautionary tale for the publishing and tech industries. The failure to verify AI-generated output before publication transforms a tool meant for efficiency into a source of misinformation.

Industry analysts note that the absolute worst application of LLMs is allowing them to perform primary work—such as sourcing quotes or verifying facts—without a human-in-the-loop verification process. When authors or researchers rely on AI as a shortcut for deep investigative work, the integrity of the final output is compromised.

The irony of The Future of Truth containing fabricated truths suggests that the risks of AI are not limited to the users Rosenbaum critiqued, but are inherent to any professional who prioritizes speed and shortcuts over traditional editorial rigor.