A group of writers, including Theranos whistleblower and Bad Blood author John Carreyrou, are suing Anthropic, Google, OpenAI, Meta, xAI, and Perplexity for training models on pirated copies of their books.
If this sounds familiar, it’s because another author has already filed a class action lawsuit against Anthropic for the same copyright infringement. In that case, the judge ruled that while it is legal for Anthropic and similar AI companies to use pirated books for training, it is not illegal to pirate the books in the first place.
Eligible writers could receive about $3,000 from the $1.5 billion Anthropic settlement, but some writers were unhappy with the resolution. The resolution does not hold AI companies accountable for the actual act of training models using stolen books that generate billions of dollars in revenue.
According to the new lawsuit, the plaintiffs say the proposed Anthropic settlement “appears to benefit[AI companies]rather than creators.”
“LLM companies should not be able to so easily wipe out thousands of high-value claims with cheap fees, ignoring what should be the true cost of intentional infringement on a large scale,” the complaint states.
