A federal judge ruled that Anthropic’s use of books to train its Claude AI model fell under fair use, comparing it to aspiring writers using existing works to innovate. However, the court declared Anthropic's storage of over 7 million pirated books infringed copyright law. A trial is set for December to determine damages, which could exceed $150,000 per work. The ruling highlights the contentious relationship between AI development and copyright protections, with implications for how creators can safeguard their intellectual property amidst the rise of generative AI technologies.
"The use of books by Anthropic to train its Claude model was deemed 'fair use' by the court, likening it to a writer's aspirational research. But their illegal storage of 7m books led to copyright infringement."
"The judge's ruling effectively underscores the complex balance between fostering innovation in AI and protecting copyright law, suggesting that transformative use is valid, but outright theft is not."
Collection
[
|
...
]