Anthropic’s “Project Panama” involved acquiring books, severing their spines, scanning the content, and pulping the paper to create a private research library for its LLM, Claude. The practice was revealed in filed documents in the northern California District Court case of Bartz v. Anthropic PBC, involving authors Andrea Bartz, Charles Graebner and Kirk Wallace Johnson, who alleged copyright infringements. The project was led by Tom Turvey, a former Google employee. An investigation by 404 Media noted the purchasing habits of AI companies, identifying ISBNdb as a broker that “gathers data from various public sources like libraries and merchants to compile a vast collection of unique book data searchable by ISBN, title, author or publisher.” ISBNdb boasts 111,978,817 searchable books and offers clients between 1,000 to 1 million books per engagement. The company advertises older print publications as “structurally guaranteed to be free of this contamination” from generative-AI. The Canadian company Zoom Books specializes in acquiring nonfiction and academic titles from the 1970s in bulk, storing titles in European warehouses, particularly in Germany, before shipping them to the U.S. and Canada for destructive scanning. Thomas Koch, press spokesperson for the German Publishers and Booksellers Association, said: “It appears to be yet another example of AI companies using vast quantities of copyright-protected works to train their language models, without consent and without payment.”