The Wikimedia Foundation, the nonprofit organization behind Wikipedia, recently urged artificial intelligence companies to stop freely scraping its vast content. Instead, the foundation wants AI developers to use its paid Application Programming Interface, Wikimedia Enterprise. This call comes as Wikipedia experiences increased server strain from automated bot traffic and an 8% decline in human page views over the past year.^{[1][2][3][4][5]} ### Why Wikipedia Needs Paid Access Wikipedia provides high-quality, human-curated information in over 300 languages, making it a valuable resource for training AI models.^[1][4] However, the surge in AI companies scraping data has created significant challenges. In May and June 2025, Wikipedia detected a large spike in automated bot traffic, with many bots attempting to evade detection by mimicking human users. This activity places a heavy burden on the nonprofit's servers.^[2][3][5][6] The rise of AI chatbots also means fewer people are visiting Wikipedia directly for answers. Human page views have dropped by 8% year-over-year.^{[2][3][4][5][6]} This shift impacts Wikipedia's primary funding model, which relies on millions of small donations from its readers. "With fewer visits to Wikipedia, fewer volunteers may grow and enrich the content, and fewer individual donors may support this work," a blog post from the Wikimedia Foundation warned.^[3][7] ### Wikimedia Enterprise API Offers Solution To address these issues, the Wikimedia Foundation is directing AI companies to its Wikimedia Enterprise API. This paid service allows organizations to access Wikipedia's content at scale without overwhelming its servers. The Enterprise API also provides structured, machine-readable data, making it easier for AI models to consume.^[1][8][2][9] Using the Wikimedia Enterprise API supports the foundation's nonprofit mission. The Wikimedia Foundation operates Wikipedia primarily through donations, with annual operating costs reaching $179 million for the 2023-2024 fiscal year.^[1][10] Wikimedia Enterprise generated $3.4 million in gross revenue in fiscal year 2023-2024, slightly up from $3.2 million the previous year.^[11] Google, for example, agreed to a deal in 2022 to commercially access Wikipedia content.^[1][6] ### Sustaining Free Knowledge Wikipedia's content is licensed under a Creative Commons Attribution-ShareAlike license, meaning it is free to use with proper attribution.^[8][12] The foundation emphasizes that AI developers should provide clear attribution to Wikipedia content contributors when incorporating information into AI outputs. This maintains transparency and respects the human effort behind the content curation.^{[2][3][13][6][7]} The Wikimedia Foundation has not threatened legal action, choosing instead a more collaborative approach.^{[2][3][6][7][14]} This differs from other content creators, like news publishers, who have filed lawsuits against AI companies for copyright infringement.^[1][12] Wikimedia also has its own AI strategy to assist its volunteer editors with tasks like translations and identifying vandalism, rather than replacing human judgment.^[2][5][6] The foundation states that human-curated knowledge is essential for accurate AI systems. Without proper recognition and financial support, the open knowledge model could be at risk.^[13] Ensuring fair compensation for data use is vital for Wikipedia to continue providing its valuable, freely accessible content to the world.^[7]