Anthropic settled copyright lawsuit over book scanning for AI training
Anthropic settled a $1.5 billion copyright lawsuit brought by book authors after court filings revealed the company had conducted a secret project to destructively scan millions of books for AI training data. The filings show Anthropic acquired and scanned books without permission, including through pirated copies, and court documents indicated other major AI companies used similar practices.
What happened
In early 2024, Anthropic executives ramped up an internal project called Project Panama, described in planning documents as "our effort to destructively scan all the books in the world," with a note stating "We don't want it to be known that we are working on this."
Within about a year, according to legal filings, Anthropic spent tens of millions of dollars to acquire and slice the spines off millions of books, then scanned their pages to feed into the AI models behind Claude.
Court filings allege that Anthropic, Meta and other companies acquired books in bulk without authors' knowledge, including by downloading pirated copies, rather than seeking direct permission from publishers and authors.
Reported impact
The practices harmed book authors whose copyrighted works were used for AI training without permission.
Index analysis
The incident reflects a pattern across major AI companies to acquire large datasets without direct authorization from copyright holders, as shown in filings across multiple copyright lawsuits against Anthropic, Meta, Google and OpenAI.
Governance controls
Anthropic paid $1.5 billion to settle the copyright lawsuit brought by book authors in August 2025.
Court documents revealed that Meta employees had raised internal concerns about the copyright violations, and the practice was approved after escalation to CEO Mark Zuckerberg.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.