Microsoft's Copilot reportedly reproduces minimal content from news articles and books, according to recent legal filings. The company stated that fewer than 1 percent of over 8 million chat logs contained significant overlaps with copyrighted material, arguing that this supports their fair use defense in ongoing lawsuits from publishers like The New York Times.
The filings come as part of a legal battle where publishers claim that Microsoft and OpenAI have built competing products using their works. Microsoft is seeking a summary judgment to potentially resolve the case early, while the outcome could have implications for how AI models utilize copyrighted content in the future.
This case highlights ongoing tensions between AI development and copyright law, particularly regarding how AI models are trained. The outcome may set important precedents for the use of copyrighted materials in technology, influencing future practices and policies in the sector.