Wikipedia wants to be the last human bastion against AI-generated content. The Wikimedia Foundation removed AI-generated summaries after several cases of hallucinations and the complaint of their editors. It’s not Wikipedia’s only problem with AI, it’s also wreaking havoc on its traffic.
What is happening. In a article published on the foundation’s blogproduct manager Marshall Miller details the current situation in Wikipedia traffic. The foundation estimates that there has been an 8% drop in human traffic between May and August 2025. They attribute this to the use of generative AI as a source of information, such as the chatbots themselves or initiatives such as the Google AI Overview that respond to the user without having to click on a link.
Why is it important. It poses a risk to the continuity of Wikipedia because people are obtaining the information that its volunteers produce, but without going through the web and without adding visits. It is the same as What’s happening in the media with Google’s AI briefingsalthough the main difference is that the media lives off advertising and Wikipedia lives off donations from individuals. Miller sees it clearly: “With fewer visits to Wikipedia, fewer volunteers will be able to develop and enrich the content, and fewer individual donors will be able to support this work.”
A crisis that comes from afar. Wikipedia losing traffic has been in the news for a long time. In 2020 they had a massive drop: They lost 3 billion organic traffic visits and the culprit was Google. The appearance of direct response modules that displayed information directly on the results page, causing many people to not click.
Correction. In May of this year, Wikipedia detected an unusual increase in traffic from Brazil. At first they classified the traffic as human, but later they verified that they were bots designed to imitate human behavior. This led them to update their bot detection mechanism and, with the new updated data, they saw that there had been an 8% drop in human traffic.
The irony. In the era of generative AI, sites like Wikipedia are an invaluable source of information. It is where the chatbots and Google’s own search draw from to give us those answers, but at the same time they are harming it and not only because of the drop in traffic, the bots and scrapers also have an impact on the operation and represent a noticeable load on the servers.
Solutions. The Wikipedia Foundation proposes responsible use by users to seek original sources and highlight the importance of human-created content. It sounds almost like a plea and the truth is that the outlook does not look too good. In the case of AI Overview, the media have warned about the consequences and there are even groups that They have sued Google. There are clues that Google could be raising licensing agreements with large media groupsbut for the moment it has not materialized and its results with AI continue to work as the first day (they have even launched the AI mode).
Image | Wikipedia
In Xataka | A Wikipedia editor spent years pretending to be Russian Red. He was actually an Indian scammer
GIPHY App Key not set. Please check settings