Authors Guild Filing Alleges OpenAI Executives Knew Data Was Pirated and Feared Negative Optics

Unsealed briefs in the Authors Guild v. OpenAI lawsuit reveal internal communications suggesting OpenAI and Microsoft executives were aware that their training data included pirated books and anticipated negative public reaction.

Internal Acknowledgment of Piracy

Documents filed in September 2026 indicate that Sam Altman and Dario Amodei disclosed the use of LibGen to Bill Gates and Microsoft CTO Kevin Scott as early as April 2019. In internal chats, researcher Sam McCandlish expressed concern that the optics of "openai uses copyrighted data from sketchy russian website" appearing on Hacker News would be "unfortunate," while Amodei described the dataset as "a bit sketchier."

Evidence Deletion and 'Project Clear'

The filings allege that OpenAI initiated a cleanup effort codenamed "Project Clear" in summer 2022 to remove LibGen files from its systems. VP of Research Bob McGrew instructed colleagues to "excise Libgen from our systems and storage," citing the company's high media profile as the reason for removing traces of the data source.

Executive Statements on Job Displacement

The briefs also highlight statements from Jack Clark, OpenAI’s Policy Director, who noted in 2020 that GPT models would "substitute for the labor" of genre fiction authors. He acknowledged that the company would likely "ignore their concerns and release anyway." Similarly, researcher Tarun Gogineni described authors' complaints about stolen datasets as "acceptable economic disruption" and mused that GPT-5 could "autocomplete" George R.R. Martin’s unfinished book series.