OpenAI partners with Condé Nast
OpenAI
OpenAI just signed a deal with Condé Nast, the company behind Vogue, Wired, and The New Yorker. ChatGPT can now surface their articles with attribution and links, not just guess at what they said.
Based on reporting by OpenAI — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
OpenAI added another big name to its growing list of media partners this week: Condé Nast. The publishing giant, home to Vogue, The New Yorker, GQ, Wired, Vanity Fair, and a dozen other titles most people recognize from a newsstand rack, is now feeding its content into ChatGPT through a licensing arrangement.
The mechanics look familiar if you've followed OpenAI's other publisher deals over the past year or so. Content from Condé Nast's outlets will show up in ChatGPT responses with clear attribution and links back to the original source, so a question about a runway trend or a tech review doesn't just produce an unsourced paraphrase. OpenAI gets training and retrieval rights to a catalog stretching back decades, and Condé Nast gets paid and, presumably, some protection against having its journalism scraped for free elsewhere.
This is part of a pattern that's been building since late 2023: Axel Springer, the Associated Press, News Corp, and now Condé Nast have all struck similar arrangements. Each deal quietly answers a question that lawsuits from The New York Times and others are asking more loudly in court — whether AI companies owe publishers anything for the material that made their models useful in the first place. OpenAI's answer, at least for the outlets willing to sign, is yes, up to a point.
What's notable about Condé Nast specifically is the breadth of its catalog. This isn't just news wire copy; it's decades of fashion criticism, investigative features, and cultural commentary that carries a distinct voice publishers worry gets flattened once it's absorbed into a chatbot's training data. Whether attribution and licensing fees are enough to keep that voice recognizable, or whether it just becomes raw material for smoother answers, is the part nobody involved wants to say out loud yet.
My take — AI-written commentary, not fact-checked reporting
These licensing deals are OpenAI buying goodwill and legal cover at the same time, and Condé Nast is smart to take the money before the negotiating leverage disappears entirely. But let's not pretend this fixes the actual tension: once your decades of reporting are training weights inside a model, attribution in a chat response doesn't return the value that got extracted. It's a nice patch, not a solution.
Read more about this at: OpenAI