Publishers are quietly cutting six-figure AI licensing deals through Snowflake’s Cortex platform, according Digiday, as the market for enterprise content licensing begins to take shape.
What do 1,000 journalists and PR pros know about AI that you don't? They took AI Quick Start, a 1-hour live class from The Media Copilot. 94% satisfaction. Find out how to work smarter with AI in just 60 minutes. Get 20% off with the code AIPRO: https://mediacopilot.ai/
The Washington Post, Associated Press, People Inc., and USA Today Network are just three of 17 publishers that have signed on to Cortex Knowledge Extensions, Snowflake’s product for connecting locked-down publisher content to enterprise AI tools via retrieval-augmented generation (RAG). Instead of scraping or exposing raw feeds, publishers can let enterprises query their paywalled or proprietary content inside Snowflake’s AI environment — and get paid for it.
Ben Srour, principal product manager at Snowflake, told Digiday the deals are real and structured in ways that make them easier for enterprises to sign. Contracts are either flat-fee licenses or usage-based, and often paid out of an enterprise customer’s existing multi-year spending commitment to Snowflake, so there’s no new procurement line required.
“You cannot scrape the data—you can’t steal it and use it for model training,” he said. “So that’s why the product has really resonated with publishers.”
Snowflake is not alone in the space. The Financial Times and The Economist have previously signaled interest in RAG royalties from opening their archives to private LLMs. AP’s chief revenue officer, Kristin Heitmann, has said the Snowflake exchange opens “unlimited use cases” covering finance companies, supply chain monitoring, crisis management, and regulatory awareness.
Snowflake’s pitch to publishers includes a point that stands out: it doesn’t take a cut of licensing deals. Snowflake makes money through storage and compute when AI queries run inside its environment. Publishers and enterprise buyers negotiate terms directly.
Snowflake also recently committed $6 billion over five years to Amazon Web Services for custom chips and AI infrastructure, a signal of how much the company is betting on AI workloads.
Not everyone is celebratory. A report from the Open Markets Institute published in April warned that AI licensing marketplaces—where AI companies pay publishers for access to articles, archives, and data—risk repeating the power imbalances of the platform era, citing the take rates platforms charge. Snowflake’s no-revenue-share model is a direct counter to that criticism.
On the enterprise side, the keenest adopters so far are financial services and marketing or communications teams already deep into building AI tools on their own data and looking for trusted external signals to plug in. Most other enterprises are still getting internal AI models and data governance in order before leaning into paid publisher content, Srour said.
Snowflake is designing Cortex for where it believes AI is going: away from chatbots and toward always-on agents quietly working in the background.
“A year ago we were talking a lot about chatbots… but very quickly things are moving into this, like, agentic, automated world,” Srour said. “What do they need? They need data. They need context. They need to know what’s happening in the world.”
