The New York Times Got Caught Publishing AI-Written Content and the Fallout Is Messy

A writer named Becky Tuch was reading a New York Times "Modern Love" column and something felt off. The writing was technically fine but had that unmistakable AI smoothness to it. She posted an excerpt on X with a careful disclaimer: "I don't want to falsely accuse writers, but this reads EXACTLY like AI slop."

Then an AI researcher named Tuhin Chakrabarty from Stony Brook University got involved. He ran the column through Pangram Labs' AI detection tool. The result? Over 60 percent of the piece was flagged as likely AI-generated.

The column's author, Kate Gilgan, responded with what might be the most 2026 answer possible. She said she did not copy and paste from AI, but she "did utilize AI as a tool" for "inspiration and guidance and correction." She listed ChatGPT, Claude, Copilot, Gemini, and Perplexity as tools she used to help stay on topic and stick to themes.

So here is the uncomfortable question: where is the line between "using AI as an editor" and "having AI write your piece"? If you are prompting five different AI tools to help you write every paragraph, and detection software flags 60 percent of the result as machine-generated, can you really call it your work?

The New York Times pointed to its ethical journalism handbook, which says AI use must "adhere to established journalistic standards." But this incident suggests the standards have not caught up with how people are actually using these tools.

This matters because if AI-written content is quietly showing up in one of the most respected newspapers in the world, it is everywhere.

As reported by The Atlantic.

Source: The Atlantic