The New York Times Testing AI Summaries on Readers
The New York Times is testing AI-generated summaries on a small group of readers, according to Semafor. This marks the paper’s first implementation of reader-facing text created entirely by artificial intelligence, without any editor’s input.
August 24, 2026 – 5:39 pm
Paris, France – December 9, 2024: A person holds an international edition of The New York Times featuring headlines about Syria, global politics, and economic issues, surrounded by a book-filled interior.
Credit: Hadrian / Shutterstock.com
The newspaper has quietly introduced a new search page that answers queries with excerpts, links to Times articles, and AI summaries of its reporting. Max Tani, Semafor’s media editor, reported this development on Sunday.
The Times aims to enhance its search feature, which Semafor describes as problematic. However, some of the paper’s journalists express alarm at this initiative.
The Times’ Statement
"We are always testing new ways for our users to discover and engage with Times journalism," said Graham James, a spokesperson for The New York Times. "This experimental feature is an example of how we are using the latest technology to create a better search experience."
Clarifying the Tool’s Capabilities (or Limitations)
It’s important to note that these summaries are generated solely from Times journalism and do not involve a chatbot answering questions about external topics. This approach has its limitations, as accurate summaries can still misrepresent the original reporting.
Previous AI Experiments
The Times has been utilizing AI in its non-news sections for several months. Wirecutter, its review site, recently introduced Wirecutter Finder, a search tool that provides AI-generated summaries and tips based on its own recommendations.
However, these internal experiments have remained separate from the newsroom, whereas the new search summaries are accessible to readers.
Union Concerns
The Times Guild, the newspaper’s editorial union, was informed about the AI summary testing during a contract bargaining session in July. According to Lachlan Cartwright’s media newsletter, Breaker, this announcement sparked frustration among union members.
The union argues that The Times is violating its own AI use standards. They have proposed AI guidelines that include an employee payment pool and human oversight of AI implementations.
The Paper’s Motives
Search and summarization tools have become standard across websites and apps. The Times aims to stay competitive and improve user experience, even as it navigates the ethical and practical challenges associated with AI integration.