Wikimedia Urges Courts to Split AI Training From Content Regurgitation in Copyright Cases

The Wikimedia Foundation is taking a middle-ground position in landmark AI copyright lawsuits, urging courts to distinguish between training large language models on copyrighted material and chatbots that regurgitate that material to users. In an amicus brief filed Friday in The New York Times' case against OpenAI and Microsoft, the nonprofit behind Wikipedia argues that AI training should qualify as fair use, while verbatim reproduction of protected works should not.

The organization takes neither side in the dispute, warning that both parties push for "overbroad and simplified applications of the fair use doctrine." A nuance-free win for publishers, it fears, could allow copyright holders to gatekeep even innovative and transformative uses of their work. Conversely, a sweeping victory for AI companies could effectively give them "a green light to snub licensing terms and fragment the flow of knowledge between human authors and audiences."

The stakes are high for Wikipedia and the broader open web. An overly restrictive ruling could deprive Wikipedians of useful AI-powered tools, while an overly permissive one could threaten the free flow of knowledge that underpins open projects. The filing highlights how even organizations sympathetic to open access see a need for courts to draw careful lines in how copyright law applies to artificial intelligence.

Read More at the original source →