In the era of generative AI, regular people's traffic on Wikipedia is declining — in 2024, an 8% increase was recorded. decrease in page views because chatbots and search engines such as Perplexity or the new Google provide answers immediately without referring to the sources. At the same time, bots and AI models consume massive amounts of Wikipedia data—often in ways that tax servers.
Jimmy Wales, the founder of Wikipedia, says it simply:
— AI companies should contribute to the costs they impose on us.
Let us remind you that at the moment Wikipedia exists mainly thanks to donations. Up to 8 million people donate subsidies to infrastructure maintenance every year.
Read also: User data in the hands of AI. Privacy is in question
Wikipedia is still a powerful resource: over 65 million articles in 300 languagesmaintained by approximately 250 thousand volunteers and the ninth most popular website in the world. The new money flowing from AI contracts is intended to help maintain servers, but also to finance the development of tools – for example, artificial intelligence tools that will support editors, for example, in removing dead links, searching for new sources, etc. Today, all this is done manually, and soon AI may help with this.
From a business perspective, this is a very smart move. Wikipedia shows that it is not fighting AI, but is becoming one of the most important suppliers of high-quality, human data for training models. This is a very valuable resource. Large language models are usually trained with random data, including from websites that offer content of questionable quality, such as X or Reddit. Not to mention that more and more AI training content is… materials previously created by AI. Access to a database of human-created content is a valuable resource for AI model creators, for which they are willing to pay.
This may also set a precedent: other open knowledge projects may also require fees from companies that download their content en masse. For Wikipedia itself, this means more stable funding and a chance to evolve — perhaps toward a more interactive, AI-integrated knowledge source that works better on phones and voice searches.
Read also: Wikipedia celebrates its fifth birthday. The website is still developing
It has already been announced, among others: improved natural language search. In practice, this means the ability to ask questions in the search field and receive not only links, but also contextual answers with links to relevant fragments of relevant articles on Wikipedia.
The risk still exists, of course. Some of the editorial community may fear that commercialization will affect neutrality or that volunteers will feel exploited. In addition, there are political controversies and competitors like Elon Musk's Grokipedia. But if Wikimedia gets it right, these deals could not only extend the life of Wikipedia, but actually strengthen its position in a world where information is becoming increasingly dominated by algorithms.
