The Wikimedia Foundation has submitted an amicus brief in the ongoing copyright infringement litigation against OpenAI. Filed on October 30, 2026, the document outlines the foundation's neutral stance, emphasizing that judicial rulings must carefully balance the interests of AI development and content creators to preserve the open web.
The foundation distinguishes between two primary AI functions. It argues that model pre-training on large datasets typically qualifies as fair use. However, it contends that retrieval-augmented generation (RAG), which directly fetches and presents specific copyrighted content, should be subject to standard licensing requirements.
Wikimedia warns that overly restrictive legal decisions could hinder technological tools that support Wikipedia contributors. Conversely, excessively permissive rulings might encourage publishers to restrict access to their content, potentially threatening the sustainability of free knowledge.
The filing highlights the scale of Wikipedia, which contains over 67 million articles in more than 300 languages. General counsel Stephen LaPorte and CEO Bernadette Meehan emphasized that the court's decision could have significant implications for the future of open access to information.
