Wikimedia says AI bots and summaries are hurting Wikipedia’s traffic

Wikimedia is sounding the alarm on the impact AI is having on reliable knowledge and information on the internet. In a , Wikimedia’s senior director of product, Marshall Miller, lays out the impact on page views that the foundation attributes to the rise of LLM chatbots and AI-generated summaries in search results.

“We believe that these declines reflect the impact of generative AI and social media on how people seek information, especially with search engines providing answers directly to searchers, often based on Wikipedia content,” said Miller.

The foundation has increasingly faced whose sophistication has made it difficult to parse human traffic from bots. After improving bot detection to yield more accurate metrics, Wikipedia’s data shows an 8 percent drop in page views year over year.

Miller paints a picture of an existential risk greater than that of a website’s page views. He posits that if Wikipedia’s traffic continues to decline, it could threaten what he calls “the only site of its scale with standards of verifiability, neutrality and transparency powering information all over the internet.” He warns that fewer visits to Wikipedia would lead to fewer volunteers, less funding and ultimately less reliable content.

The solution he offers is for LLMs and search results to be more intentional in giving users the opportunity to interact directly with the source for the information being presented. “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,” Miller writes.

Earlier this summer, Wikipedia floated the idea of AI-generated summaries that would appear at the top of articles. The project was before it began after fierce backlash from the site’s volunteer editors.