Meta Signs $150M Deal to License News Corp Content for AI

Meta has signed a three-year deal worth up to $150 million to license content from News Corp for AI model training and Meta AI chatbot responses, Dataconomy reported. The agreement covers the Wall Street Journal, New York Post, The Australian, The Times (UK), and other News Corp properties at roughly $50 million per year.

Deal Terms

The agreement allows Meta to use News Corp's editorial content for two purposes: training data for its AI models (including the Llama family) and as a source for Meta AI chatbot responses across Facebook, Instagram, WhatsApp, and the standalone Meta AI app.

At $50 million per year, this is the second-largest known AI content licensing deal for a single publisher, behind News Corp's own $250 million agreement with OpenAI signed in May 2024. Combined, News Corp is now earning roughly $133 million per year from AI licensing alone - more than many of its individual newspaper properties produce in subscription revenue.

The "Input Company" Framing

News Corp CEO Robert Thomson described his company as an "input company" for AI during the announcement - a framing that represents a significant strategic shift for the media industry. Rather than fighting AI companies over unauthorized training data use (as the New York Times has done through its lawsuit against OpenAI), News Corp is positioning itself as a premium supplier.

The logic: if AI models are going to use news content regardless, it's better to license it at a price that reflects its value than to litigate after the fact. News Corp gets guaranteed revenue. Meta gets legal certainty and premium content that may improve its models' factual accuracy on current events.

The Broader Licensing Landscape

The Meta deal extends a pattern that began with OpenAI's publisher partnerships in 2024:

Not every publisher is licensing. The New York Times, Washington Post, and several European publishers continue to pursue litigation rather than deals. The split reflects a fundamental disagreement about whether AI training on published content is fair use or infringement - a question no court has definitively resolved.

For Meta specifically, the deal addresses a competitive gap. OpenAI's ChatGPT already cites sources from its licensed publishers. Google's Gemini draws on its search index. Meta AI has had weaker access to premium news content, which affects the quality of its responses on current events and factual queries. $50 million per year buys access to some of the most-cited business journalism in the English-speaking world.

The economics are simple enough. News Corp's total AI licensing revenue now passes $400 million across the OpenAI and Meta deals. That is more than what most individual newspapers earn in a year. Thomson's "input company" framing may sound like surrender to critics of AI training on journalism, but the revenue tells a different story - at least for publishers large enough to negotiate from a position of leverage.

Sources:

• Meta Signs $150 Million AI Content Deal with News Corp - Dataconomy
• Meta Signs Multimillion-Dollar AI Licensing Deal with News Corp - Engadget
• News Corp and OpenAI Sign Content Deal - WSJ