Wikimedia urges nuance in AI copyright case between publishers and tech firms

Summarised from 2 outlets · Updated 30 Oct, 17:11 · Archive

The Wikimedia Foundation has filed a legal brief in the New York Times' copyright lawsuit against OpenAI and Microsoft, seeking a middle ground between the opposing positions in the landmark case. The nonprofit warns that an overly black-and-white ruling could either deprive Wikipedia editors of useful AI tools or endanger the open web by allowing tech companies to ignore licensing terms.

Wikimedia argues that large language models trained on copyrighted work should receive legal protection as a fair use of that content. However, it opposes chatbots that reproduce copyrighted text directly to users without adhering to the terms under which the material was originally licensed. The organization notes that Wikipedia has become a critical source of data for AI training, but these same systems undermine the communities that maintain Wikipedia by siphoning away visitors.

Wikimedia's brief proposes protecting certain AI training scenarios and a technique called retrieval-augmented generation, which optimizes AI outputs by referencing external knowledge bases. However, it does not support retrieval-augmented generation when large language models regurgitate pre-existing articles without attribution. The organization says that AI systems could help Wikipedia editors by detecting copyright violations, fighting vandalism and identifying articles needing improvement.

How it is being reported

• The Wikimedia Foundation weighs in on the NYT's case against OpenAIThe folks behind Wikipedia say that an extreme decision in either direction could threaten the open web.Engadget · 30 Oct, 15:47
• Wikipedia’s operator says a bad AI copyright ruling could threaten the open webThe Wikimedia Foundation is staking out a middle ground in a group of landmark AI copyright cases, pushing courts to legally protect large language models (LLMs) training on copyrighted work - but not chatbots regurgitating that work to users. An overly black-and-white ruling, it says, could either deprive Wikipedians of useful AI-based tools or endanger […]The Verge · 30 Oct, 09:00

In this story: Wikimedia Foundation · Wikipedia · The New York Times · OpenAI · Microsoft

This summary was written by AI from the headlines and standfirsts above, and states as fact only what at least two outlets report. How we use AI · Report a problem · Corrections log