Meta is reversing course on news after years of deprioritising journalism. The social media giant has signed multiple content licensing deals with major publishers worldwide, acknowledging that its artificial intelligence system struggles with current events and real-world accuracy.
The most significant agreement is a three-year deal with News Corp valued at up to $50 million annually. The agreement will allow Meta to use News Corp's news content, including archives, to retrieve real-time information about current events for its AI products and also to train its AI models. News Corp properties include The Wall Street Journal, New York Post, Times of London, and The Australian.
Beyond News Corp, Meta is partnering with Le Figaro, Prisa and Süddeutsche Zeitung, adding French, Spanish, and German publishers to the mix. Earlier deals brought USA Today, People Inc., CNN and Fox News into Meta's content network. When you ask Meta AI news-related questions, you'll now receive information and links that draw from more diverse content sources to help you discover timely and relevant content tailored to your interests.
The urgency reflects a practical problem. Meta AI has demonstrated significant accuracy gaps with current information. The system has repeatedly failed on straightforward factual questions, a limitation that becomes critical as users expect AI assistants to handle real-time queries about breaking news and current events.
This pivot represents a sharp departure from Meta's recent strategy. The new publishing deals mark a shift from Meta's recent decisions to reduce the role of news on its platforms. The company shut down the Facebook News tab in 2024 and previously ended payments to news publishers in 2022. CEO Mark Zuckerberg had spent years dephasising news content in user feeds following misinformation controversies. Now, with Meta struggling to compete with ChatGPT and Google Gemini, the company sees news content as essential infrastructure for its AI ambitions.
Publishers face a mixed proposition. Meta gets credible, timely information to address AI hallucination concerns, while publishers gain distribution to the company's massive user base spanning Facebook, Instagram, WhatsApp and beyond. Yet the industry observers note risks. Publishers get distribution but they're also training the systems that could eventually replace them. News organisations watched Google and Meta hollow out their advertising businesses over the past two decades, and now they're handing over their content to AI companies with unclear long-term implications.
The deals reflect broader competitive dynamics across Big Tech. OpenAI and Google have been racing to ink similar deals with publishers, recognising that large language models trained on static datasets struggle with anything happening right now. News Corp has already secured a separate five-year deal with OpenAI reportedly worth $250 million, giving the publisher leverage to negotiate with Meta. News Corp CEO Robert Thomson said he expected the value of the company's assets to increase over time as multiple sectors seek out reliable data to boost their products.
The arrangement does bring publishers tangible benefits. Linking out to articles allows publishers to visit these partners' websites for more details while providing value to partners, enabling them to reach new audiences. For financially struggling news organisations, consistent licensing payments offer more reliable income than eroding digital advertising revenue.
Yet important questions remain unresolved. It's less clear how Meta will handle the thorny moderation questions that inevitably come with news content. How will it handle breaking news that's still developing? Meta's announcement did not address how its system will treat disputed information or articles containing errors that have yet to be corrected.
The competitive calculus for smaller publishers is harsher. Unlike News Corp, which can negotiate from a position of institutional strength, regional and independent outlets lack the bargaining power to secure favourable terms. Smaller publishers face tougher choices with limited legal resources and less negotiating power. This creates a structural risk where only the largest media companies can secure the payments needed to survive an AI-driven future.