The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →The New York Times ended its relationship with freelance book reviewer Alex Preston after it found that his review of Watching Over Her contained language resembling an earlier Guardian review. The Times said Preston had used an AI tool to help draft the review and failed to catch the overlap; Preston described it as improper use of an AI editing tool on a draft he had written. The episode also sharpened a dispute between Times management and its union over how AI use should be governed and disclosed.
What happened with Alex Preston’s New York Times review?
The case concerned one review, not a finding about all of Preston’s work. The review of Jean-Baptiste Andrea’s novel Watching Over Her was published by The Times on Jan. 6. A reader later noticed similarities to a Guardian review of the same book published in August, according to TheWrap’s March 30, 2026 report.
The Times investigated and spoke with Preston. TheWrap reported that he acknowledged using an AI tool to help draft the piece and said he had not used AI to help draft his other stories. The Times added an editors’ note identifying the AI use and linking to the Guardian review, notified the Guardian about the similarities, and said Preston would no longer write for the paper. TheWrap reported that he had written six Times reviews between 2021 and 2026.
Preston told TheWrap, “I used an ‘A.I. editing tool improperly on a draft I had written’” and said he failed to catch “overlapping language” from the Guardian review. He also said, “I took responsibility immediately and apologized to The New York Times.” Those are Preston’s descriptions of what happened.
#1 Best Overall
A Times spokesperson characterized the matter as a standards violation: “For staff journalists and freelance writers alike, reliance on A.I. and inclusion of unattributed work by another writer is a serious violation of The Times’s integrity and fundamental journalistic standards.” Preston was a freelancer, not a Times staff employee.
What the incident does—and does not—show
The accounts support a straightforward conclusion: The Times attributed the similarities to AI-assisted work and Preston’s failure to catch them, while Preston accepted responsibility for using an AI editing tool improperly. The available reporting does not establish the technical process by which the tool produced similar wording. It does not say whether the cause involved retrieval, prompting, training data, or another mechanism, so assigning a technical cause would go beyond what is known.
Rank #2
- Ideal for Gifting
- Ideal for a bookworm
- Compact for travelling
- The reported finding concerns this review; it is not evidence that every AI-assisted contribution contains copied language.
- The overlap was noticed by a reader. The reporting does not say an AI detector discovered it.
- The Times’ stated concern was both reliance on AI and unattributed work resembling another writer’s work—not simply that an AI tool was used.
Why the case became part of a larger AI policy dispute
The Preston case unfolded amid contract negotiations between The Times and the NewsGuild of New York, which represents Times employees. The Guild’s AI subcommittee argued that the company’s standards were unclear and sought stronger protections. Axios reported that Guild members wanted clearer disclosure rules, safeguards for AI use in performance reviews, and protections for workers’ names, images, and likenesses. The Guild letter described it as representing 1,500 members in Axios’s April 7, 2026 coverage.
In an April 8 follow-up, TheWrap reported that the Guild viewed public guidance as unclear or open to interpretation and wanted the company to state more clearly how AI is used in stories. Times management said it supported strong standards but resisted putting detailed restrictions into a multi-year contract because the technology changes quickly. Managing editor Marc Lacey wrote, “AI technology is ceaselessly evolving – quickly – and we believe that this rapid change is precisely why we must remain flexible.”
The disagreement was still live in the April reporting; it does not establish that the parties reached a settlement. Their positions expose several distinct policy questions:
| Question | Guild’s reported concern | Management’s reported position |
|---|---|---|
| Disclosure | Make it clearer when and how AI is used in stories. | Supports strong standards; detailed contractual requirements were not specified in the report. |
| Permitted assistance | Seek clearer limits and rules around AI use. | Wants flexibility as the technology changes; the reporting does not establish a final drafting-versus-editing rule. |
| Human accountability | Protect trust in published work and clarify standards. | Supports strong standards; the reports do not describe a settled accountability framework. |
| Worker protections | Address AI use in performance reviews and protect names, images, and likenesses. | The cited reports do not describe a management proposal on those specific protections. |
| How rules are set | Seek firmer protections through contract talks. | Avoid locking detailed restrictions into a multi-year contract because AI is changing rapidly. |
The table reflects the positions reported during negotiations, not agreed policy language. TheWrap also quoted the Guild’s AI subcommittee saying, “When The Times instead publishes A.I.-generated work, intentionally or not, our readers lose trust in what we do. This is unacceptable.”
Rank #4
Why AI detector scores cannot prove what happened
The Preston review should not be treated as a detector case: the reporting says a reader raised the similarities, not that software identified them. More broadly, AI-text detectors can disagree, miss AI-generated material, and produce false positives. A score alone cannot establish who wrote a passage or whether it was copied.
A separate example illustrates the problem without being evidence about Preston. The Atlantic reported on Modern Love contributor Kate Gilgan’s acknowledged use of AI for inspiration, guidance, and correction; Gilgan said, “I used AI as a collaborative editor and not as a content generator.” In that separate column, detector estimates varied: Pangram estimated more than 60 percent AI-generated, two other tools flagged 30 percent as likely AI-generated, one found no AI, and another suspected AI without giving a percentage. The Atlantic cautioned that such estimates are difficult to establish with certainty. These were tools’ estimates about Gilgan’s column, not validated authorship measurements and not results for Preston’s review. See The Atlantic’s March 25, 2026 report.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsBest Value
- It can be a gift option
- Comes with secure packaging
- Helpful in various ways
What readers should take away
The immediate outcome was that The Times added an editors’ note and ended its freelance relationship with Preston after a review contained language resembling an earlier Guardian piece. The Times framed the issue as a breach involving AI reliance and unattributed work; Preston said he had used an AI editing tool improperly and failed to catch the overlap. The broader policy question—how to disclose, limit, and oversee AI use while rules adapt to changing tools—remained unresolved in the April coverage.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
