Wikipedia is seeking new licensing deals with AI companies as rising automated scraping strains the nonprofit’s infrastructure and funding model.
 |
As AI companies rely heavily on Wikipedia data, the nonprofit seeks licensing deals to cover escalating operational costs and ensure sustainability. Image: Wikimedia/CH |
Tech Desk — December 4, 2025:
Wikipedia is moving to renegotiate its relationship with the artificial intelligence industry as AI developers increasingly depend on the platform’s vast information store to train large language models. Speaking at the Reuters NEXT summit in New York, co-founder Jimmy Wales said the nonprofit is pursuing additional licensing agreements similar to its existing arrangement with Google in an effort to offset the growing financial burden created by automated scraping.
While Wikipedia’s content remains free for public use, Wales emphasized that large-scale crawler activity from AI companies places disproportionate strain on the Wikimedia Foundation’s infrastructure. “The AI bots that are crawling Wikipedia are going across the entirety of the site… so we have to have more servers, we have to have more RAM and memory for caching that, and that costs us a disproportionate amount,” he said.
Google, which signed a licensing deal in 2022, is currently the model for what Wales hopes to replicate with other companies, including those whose AI systems rely heavily on Wikipedia for training data. Although Wikipedia’s open-access mission has long supported the free flow of information, Wales stressed that the foundation’s donation-driven funding model was never meant to subsidize multibillion-dollar AI development. “Those people are donating money to support Wikipedia, and not to subsidize OpenAI costing us a ton of money,” he said.
The push for licensing raises broader questions about whether AI companies should compensate public and nonprofit organizations that provide the raw materials behind modern AI tools. With Wikipedia serving as the world’s largest open knowledge repository, the debate highlights emerging tensions between open-access ideals and the commercial realities of AI development at scale.
Pressed on the possibility of legal action against AI firms that use Wikipedia without paying, Wales said he was unsure but noted the platform’s “soft power” and ability to publicly pressure companies could be an effective alternative. He added that technical measures, such as Cloudflare’s AI Crawl Control, might be considered to moderate scraping—but acknowledged such restrictions could conflict with Wikipedia’s longstanding commitment to openness.
For more than two decades, the Wikimedia Foundation has operated Wikipedia through volunteer editors and public contributions while striving to maintain neutrality, even during global crises. Wales expressed confidence in the community’s consistent stewardship, saying: “The community tends to do a pretty good job, even with those circumstances.”
Wikipedia’s evolving stance toward AI firms reflects a larger shift in how public knowledge infrastructures navigate the demands of a rapidly expanding AI ecosystem—one that increasingly relies on them, yet often contributes little to their sustainability.