Wikipedia urges AI companies to use its paid API, and stop scraping

Wikipedia has outlined a plan to ensure its website remains supported in the AI era, even as it faces declining traffic. In a recent blog post, the Wikimedia Foundation, which operates the online encyclopedia, called on AI developers to use its content responsibly. The organization asks developers to ensure proper attribution for its contributions and to access content through its paid product, the Wikimedia Enterprise platform.

This optional, paid product enables companies to use Wikipedia’s content at scale without placing a severe strain on its servers. The paid nature of the service also allows AI companies to support the organization’s nonprofit mission.

While the post did not threaten penalties or legal action for unauthorized scraping, Wikipedia recently identified that AI bots had been scraping its website while attempting to appear human. After updating its bot detection systems, the organization discovered that unusually high traffic in May and June came from AI bots trying to evade detection. It also reported that human page views had declined by eight percent year-over-year.

Wikipedia is now establishing guidelines for AI developers. It states that generative AI developers should provide attribution to credit the human contributors whose content is used to create AI outputs. The post explains that for people to trust information online, platforms must clarify where information is sourced from and encourage visits to those sources. It warns that with fewer visits to Wikipedia, fewer volunteers may grow and enrich the content, and fewer individual donors may support the work.

Earlier this year, the organization released an AI strategy for its editors. This strategy focuses on using AI to assist editors with workflows, including automating tedious tasks and translation, emphasizing that these tools are meant to help editors, not replace them.