In recent rulings, judges have granted favorable decisions to Meta and Anthropic on their use of copyrighted material for AI training, labeling it as 'transformative' fair use. However, Judge Chhabria expressed his reservations about the arguments made by authors opposing Meta, stating they failed to demonstrate harm to their market. His ruling emphasizes deficiencies in the plaintiffs' case regarding the impact of AI-generated content, showcasing a complex legal landscape where AI training practices are being contested in court without a clear consensus.
This ruling does not stand for the proposition that Meta's use of copyrighted materials to train its language models is lawful, but only that these plaintiffs made wrong arguments.
The authors fatally only argued that users of Llama can reproduce text from their books, rather than addressing how it could harm their market.
Chhabria found the authors' theories flawed, particularly their claim that Llama could produce long excerpts from their works.
Ultimately, the absence of evidence showing that Meta's AI threatens to dilute the authors' markets led to Chhabria granting summary judgment in favor of Meta.
Collection
[
|
...
]