Science & Tech · AI · 2 hrs ago
Wikimedia urges middle ground in AI copyright case
The Wikimedia Foundation has urged a middle ground in a US copyright case brought by The New York Times against OpenAI and Microsoft.
The case asks whether the companies violated copyright by using Times articles to train AI systems and by reproducing parts of them.
Wikimedia says training can be fair use when models create something new, rather than simply copying their sources.
But it says chatbots should need permission to reproduce works in answers, and should respect licensing terms and credit original authors.
The foundation warns that ruling too strictly against AI training could limit useful tools, while allowing any use could weaken the open web and its sources of knowledge.
That concern affects Wikipedia directly: its content is used to train AI, while chatbots may draw visitors away and reproduce licensed text without meeting its terms.
A court decision is expected soon and could influence other AI copyright cases.
The Wikimedia Foundation has urged a middle ground in copyright cases involving The New York Times, OpenAI and Microsoft.
In an amicus brief filed in a US court, it argued that using copyrighted works to train large language models can qualify as fair use when the models produce something new and different.
It said AI companies should respect licenses when chatbots reproduce copyrighted material in responses to users.
The foundation warned that either an overly restrictive or blanket fair-use ruling could harm innovation and the open web.
- Who
- The Wikimedia Foundation, which operates Wikipedia, filed the brief in cases involving The New York Times, OpenAI and Microsoft.
- What
- It urged a middle ground on whether AI companies’ use of copyrighted material is fair use.
- When
- The brief was filed on Friday, October 30, 2026.
- Where
- In a US court.
- Why
- The foundation said an overly restrictive ruling could inhibit innovation, while blanket fair use could weaken access to online knowledge and disregard licensing terms.
The New York Times
OpenAI and Microsoft
Use of copyrighted articles
The New York Times
The Times says companies should not have free rein to use its published content.
OpenAI and Microsoft
They argue that training and deploying AI models on published data is necessary for technological development and can qualify as fair use.
The New York Times filed its lawsuit against OpenAI and Microsoft, alleging copyright infringement.
The Wikimedia Foundation filed an amicus brief urging a middle ground on AI use of copyrighted works.
A summary judgment in the lawsuit is expected, according to Engadget.
- Brief
- Amicus brief filed by the Wikimedia Foundation
- Filing date
- October 30, 2026
- Parties in the Times case
- The New York Times, OpenAI and Microsoft
- Proposed training standard
- Fair use when models produce something new and different from the original sources
- Retrieval-augmented generation
- The foundation says licensing terms should be respected when copyrighted material is reproduced for users