In a move that underscores the growing tension between traditional journalistic integrity and the convenience of generative technology, The New York Times officially cut ties with freelance journalist and novelist Alex Preston on March 30, 2026. The decision followed an internal investigation which revealed that Preston had used an artificial intelligence tool to assist in drafting a book review, a tool that inadvertently “lifted” verbatim passages from a rival publication.
The incident has sent shockwaves through the literary and media communities, serving as a stark warning about the invisible pitfalls of “AI-assisted” writing in an era where transparency is the ultimate currency.
The “Smoking Gun”: A Reader’s Discovery
The controversy began in late March when an eagle-eyed reader noticed striking similarities between a review published in The New York Times on January 6, 2026, and an earlier piece from The Guardian. The subject was Jean-Baptiste Andrea’s acclaimed novel, Watching Over Her.
Upon closer inspection, the reader found that Preston’s review for the Times contained descriptions of characters and thematic summaries that were nearly identical to those written by Christobel Kent in The Guardian back in August 2025. Phrases like “lazy, Machiavellian Stefano” and specific closing metaphors about Italy as a “song of performance” appeared in both texts with only minor punctuation differences. The reader alerted the Times editorial board, prompting an immediate inquiry into the freelancer’s process.
The AI Trap: How “Assistance” Became Plagiarism
When confronted by Times editors, Preston was reportedly forthright about his mistake. He admitted to using an AI editing tool to help structure his draft and refine his prose. However, the tool he employed likely utilizing a Large Language Model (LLM)—did more than just offer synonyms; it seemingly “crawled” existing reviews of the same book to supplement the draft.
Preston claimed he failed to identify that the AI had incorporated specific sentences from Kent’s Guardian review into his final submission. This phenomenon, often called “stochastic parroting,” occurs when AI models regurgitate training data or web-scraped content without attribution. For Preston, a seasoned author with six books to his name, the failure to vet the “AI-generated” sections resulted in a career-defining breach of trust.
The NYT’s Hard Line on Journalistic Integrity
The response from The New York Times was swift and uncompromising. A spokesperson for the paper described the incident as a “serious violation of integrity and fundamental journalistic standards.” For a legacy institution currently embroiled in high-stakes legal battles with AI companies like OpenAI over copyright and data scraping, the optics of its own contributors using those same tools to plagiarize were intolerable.
The Times took the following actions:
-
Permanent Severance: The paper confirmed that Preston will no longer write for any of its sections.
-
Editorial Transparency: An extensive editor’s note was appended to the online version of the January review, acknowledging the use of AI and linking directly to the original Guardian piece.
-
Audit of Past Work: While Preston asserted that this was the only instance in which he used AI in his six reviews for the paper since 2021, the Times conducted a retrospective audit of his previous work to ensure no other violations occurred.
The Human Element: Alex Preston’s Public Apology
Alex Preston, whose most recent book A Stranger in Corfu was published just last month, expressed deep remorse following the scandal. In a statement to The Guardian, he described himself as “hugely embarrassed” and admitted to making a “serious mistake.”
“I used an AI tool improperly on a draft I had written and missed the overlapping language from another review that the AI dropped in,” Preston said. “I took responsibility immediately and apologized to The New York Times, Christobel Kent, and The Guardian.”
The fallout is particularly ironic given Preston’s professional background; beyond his literary career, he serves as a senior advisor for an investment management firm and had recently written an essay titled The AI Bubble: Hidden Risks and Opportunities. The “hidden risk” he warned others about ultimately became his own professional undoing.
The “New Normal” for Editorial Guardrails
This incident marks a turning point for how media organizations manage freelancers in 2026. As AI tools become embedded in word processors and search engines, the line between “editing help” and “content generation” has blurred.
Industry analysts suggest that the Preston case will force major outlets to implement stricter “AI-detection” protocols and mandatory disclosure agreements for all contributors. For the Times, the message is clear: while the paper may be experimenting with AI in its business operations, the “human signal” of its criticism must remain untainted. In the world of high-stakes journalism, the cost of a “shortcut” has never been higher.
