The New York Times Has Been Publishing AI-Written Content and Hoping Nobody Would Notice

Here's something that might make you look at your morning news differently.

A writer named Becky Tuch was reading a New York Times "Modern Love" column when something felt off. The writing was technically fine, but it had that weird, too-polished quality that screams "a robot wrote this." She posted about it on X, and it blew up.

Then an AI researcher from Stony Brook University, Tuhin Chakrabarty, ran the entire column through an AI detection tool. The result? Over 60% of the piece was flagged as likely AI-generated. Two more detection tools flagged about 30% of it as AI.

When confronted, the author admitted she used ChatGPT, Claude, Gemini, Copilot, and Perplexity while writing the piece. She called it "collaborative editing" rather than content generation. But here's the thing: when you're using five different AI tools to write one essay, where exactly does your voice end and the machine begin?

The New York Times says its handbook requires freelancers to follow ethical journalism standards around AI use, but clearly something slipped through the cracks. And if it happened once in a high-profile column, how often is it happening everywhere else?

As reported by The Atlantic.

Source: The Atlantic