The Wikimedia Foundation signed licensing deals with Microsoft, Meta, and Amazon, marking the first time the nonprofit has formally monetized access to the volunteer-written encyclopedia that quietly powers much of the internet's AI training data.
Paid API access replaces the scraping that tech companies have been doing for years. The deals include attribution requirements and structured data feeds, though exact pricing remains undisclosed.
Revenue beyond donations gives Wikimedia a new funding stream as AI companies race to lock up training data. Wikipedia has been a foundational dataset for language models since GPT-2, but the organization never saw a dollar from it.
No exclusive rights in the deal. All Wikipedia content stays freely available under Creative Commons. The companies are essentially paying for enterprise-grade API access and support, not content lockup.
Editor community backlash is the wild card. Wikipedia's volunteer base has historically resisted any whiff of commercialization, and some editors have already voiced concerns about the nonprofit profiting from their unpaid work while Big Tech builds billion-dollar products.
The deals signal a broader shift in how AI companies acquire training data. Free scraping is giving way to paid licensing, and content owners are finally getting a seat at the table.
WorldClaw, a new paper from Tencent Hunyuan, describes a text-to-3D system that builds large, explorable scenes and returns editable, instance-level assets rather than a single locked render. The method leans on plann…
Matriarch, the self-funded Los Angeles studio from Verena Puhm, is developing its own audience-response technology alongside its film and series slate rather than adding it to a traditional studio later. Puhm told Dea…
LTX-2.5 generates connected shots that hold character, environment, lighting, and voice across cuts, then outputs native 4K HDR and a RAW workflow for professional color and finishing. The open-weights model runs on h…
