Wikipedia will share content with AI firms in new licensing deals

In a significant move, the Wikimedia Foundation has unveiled new licensing agreements with prominent tech companies including Microsoft, Meta, Amazon, Perplexity, and Mistral AI. This initiative aims to monetize the use of Wikipedia content, which has been a key resource for training AI models that fuel popular AI assistants like Microsoft Copilot and OpenAI’s ChatGPT. Historically, many of these companies accessed Wikipedia data without authorization. However, this new arrangement indicates that the majority of leading AI developers are now participants in Wikimedia's Enterprise program. This commercial subsidiary offers API access to Wikipedia’s vast repository of 65 million articles, enabling faster and more extensive data retrieval than what is available through free public APIs. While the Wikimedia Foundation has not disclosed the financial specifics of these agreements, the inclusion of these new partners is noteworthy. They join Google, which entered into a similar agreement with Wikimedia Enterprise back in 2022, along with smaller firms such as Ecosia, Nomic, Pleias, ProRata, and Reef Media. The revenue generated from these deals is crucial for covering the nonprofit's infrastructure costs, which would otherwise depend on modest public donations, particularly as its content becomes increasingly integral to AI training datasets. Lane Becker, the president of Wikimedia Enterprise, emphasized the importance of financial support from these tech firms. He stated, “Wikipedia is a critical component of these tech companies’ work that they need to figure out how to support financially.” Becker acknowledged that it took time to develop the right features to encourage these companies to transition from free access to a commercial model, but he noted that all major partners recognize the necessity of contributing to the sustainability of Wikipedia’s mission.