Wikipedia chooses the middle ground in New York Times and OpenAI lawsuit

The top 3

The proposed technical distinction in fair use: The Wikimedia Foundation argues that restricting AI training too heavily would stifle open-source innovation, but permitting 'blanket' fair use could destroy the economic ecosystem that produces the information AI relies on. By separating the input stage (model training) from the output stage (RAG or chatbot responses), they aim to protect both technological progress and the secondary sources that contribute to collective knowledge.
Wikipedia's dual-incentive landscape: Wikipedia balances the benefits of AI in assisting human editors (such as detecting plagiarism and vandalism) against the risks of traffic cannibalization and copyright non-compliance. Their intervention in the New York Times v. OpenAI case highlights a concern that if information sources are not compensated or protected, the web will become more closed, ultimately degrading the quality of free information available to LLMs.