Wikipedia Reports 8% Drop in Human Pageviews Amid AI Search Growth

Oct 20, 2025
The Wikimedia Foundation says Wikipedia’s human pageviews have fallen 8% year-over-year, citing the impact of AI-powered search results and social video platforms on how people access information.

Wikipedia’s human pageviews have declined by about 8% compared to last year, according to a post on the Wikimedia Foundation’s Diff blog. The organization attributes the drop to changes in how people seek information, including the rise of AI-generated search summaries and the popularity of social video platforms.

The foundation recently updated its bot detection systems after discovering unusually high traffic in May and June 2025 that was actually coming from bots disguised as human users. After reclassifying the data, the analysis showed a clear decrease in genuine human visits across all language versions of Wikipedia.

Marshall Miller of the Wikimedia Foundation noted that search engines increasingly provide direct answers using Wikipedia-sourced information rather than linking to the site itself. He added that younger audiences are turning to platforms like TikTok and YouTube for information instead of visiting traditional websites.

In response, the foundation is developing a framework for proper attribution of its content and encouraging AI and search companies to direct more visitors to Wikipedia. It has also formed two new teams focused on reader growth and experience, and is experimenting with ways to reach younger users through videos, games, and chatbots.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more
Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.