Wikipedia is trying to set clearer terms for how AI developers use the encyclopedia’s content. The Wikimedia Foundation, which runs the site, says AI companies should access Wikipedia responsibly, credit the people behind the information, and use its paid Wikimedia Enterprise platform instead of scraping pages at scale.
Wikipedia’s message to AI developers
In a blog post, the Wikimedia Foundation called on AI developers to use Wikipedia content in ways that preserve the health of the project. The request centers on two ideas: proper attribution and responsible technical access.
The foundation wants generative AI developers and providers to make clear when their outputs rely on Wikipedia’s work. That means crediting the human contributors whose edits, research, and maintenance make the online encyclopedia useful in the first place.
It also wants companies that need Wikipedia content at scale to use Wikimedia Enterprise. That product is opt-in and paid, and it is designed for organizations that rely on large amounts of Wikipedia material.
The foundation’s position is not simply that AI companies should pay for access. It is also arguing that the way content is collected matters. According to the source article, Wikimedia Enterprise lets companies use Wikipedia’s content without “severely taxing Wikipedia’s servers.”
Why scraping has become a bigger concern
The Wikimedia Foundation is making this push while traffic patterns around Wikipedia are changing. The organization recently noted that AI bots had been scraping its site while trying to appear human.
After an update to its bot-detection systems, the organization found that unusually high traffic in May and June came from AI bots attempting to “evade detection.” At the same time, it said that “human page views” had declined 8% year-over-year.
Those two details help explain the foundation’s urgency. If bots can create heavy traffic while actual human visits decline, Wikipedia faces two pressures at once: infrastructure strain from automated access and a weaker relationship with readers who might otherwise become volunteers or donors.
The blog post, as described in the source article, does not threaten penalties or legal action over scraping. Instead, the Wikimedia Foundation is laying out expectations for AI developers: use the paid channel built for scaled access, and show users where information comes from.
Attribution is part of the trust problem
For Wikipedia, attribution is not only about giving credit. It is also about helping people understand the source of the information they receive online.
The foundation wrote: “For people to trust information shared on the internet, platforms should make it clear where the information is sourced from and elevate opportunities to visit and participate in those sources.” That statement connects AI output to the broader information ecosystem.
If an AI product uses Wikipedia content without pointing people back to Wikipedia, the user may get an answer but never see the underlying community that produced and maintains the source material. The foundation’s concern is that this can weaken the pathways that keep Wikipedia active.
The blog post also warns: “With fewer visits to Wikipedia, fewer volunteers may grow and enrich the content, and fewer individual donors may support this work.” That is the core sustainability issue. Wikipedia depends on participation, and participation is harder to grow when fewer people reach the site directly.
What Wikimedia Enterprise is meant to solve
Wikimedia Enterprise gives companies a structured way to use Wikipedia content at scale. The source article describes it as an opt-in, paid product that supports large-scale use while reducing the burden on Wikipedia’s servers.
For AI companies, the platform offers a more direct route to Wikipedia content than scraping. For the Wikimedia Foundation, it creates a way for heavy commercial users to support the organization’s nonprofit mission.
The distinction matters because Wikipedia is not just a database. It is a live project maintained by people. The foundation’s argument is that large AI systems should not separate the content from the community and infrastructure that make the content possible.
In plain terms, the request gives AI developers a clearer checklist:
- Use Wikimedia Enterprise when accessing Wikipedia content at scale.
- Provide attribution when Wikipedia content helps produce AI outputs.
- Avoid scraping practices that strain servers or attempt to evade detection.
- Help users find and participate in the original sources of information.
How Wikipedia is approaching AI internally
The Wikimedia Foundation is not rejecting AI outright. Earlier this year, it released its AI strategy for editors, which described using AI to support editorial workflows.
That strategy focused on helping editors with tedious tasks, automating translation, and building tools that assist editors rather than replace them. This detail is important because it shows the foundation is drawing a line between AI as support for human contributors and AI as a system that extracts value from their work without reinforcing the source.
The broader message is consistent: AI can be useful when it strengthens the people and processes behind Wikipedia. It becomes a problem when it obscures attribution, reduces direct engagement, or puts technical pressure on the site without supporting the project.
For AI developers, Wikipedia’s request is a signal that widely used public-interest resources are starting to define their expectations more clearly. The foundation is not framing the issue only as access to content. It is framing it as a question of trust, attribution, server load, volunteer growth, and long-term support for a nonprofit knowledge project.
As generative AI products rely on information from the open web, Wikipedia’s position highlights a central tension: AI systems need high-quality sources, but those sources still need visitors, contributors, and funding to remain useful. The Wikimedia Foundation is now asking AI companies to acknowledge that dependency more directly.