Meta’s effort to shift blame for AI-training copyright violations onto employees ran into a discovery order in litigation involving authors Richard Kadrey and Sarah Silverman and publisher Entrepreneur Media. Magistrate Judge Thomas Hixon largely agreed that Meta must identify executives and other supervisors who approved downloading books from shadow libraries including Anna’s Archive and LibGen, following testimony from employees who performed the downloads. Meta had characterized its early dataset-building process as informal and lacking clear decision-makers, arguing that a decision occurred only in the technical sense that workers downloaded torrents; the order could expose a broader chain of command. The disputed acquisitions totaled 80.6 terabytes from LibGen and 35.7 terabytes through Anna’s Archive. Authors argue that Meta deliberately induced infringement and that obtaining data at that scale required seeding, contrary to the company’s claimed use of anti-seeding “mitigations.” Internal messages and testimony show that employees objected to torrenting from corporate laptops, one download required director- or vice-president-level approval, and Anna’s Archive downloads were halted before leadership resumed them. Meta has sought to withhold communications identifying approvers as privileged, but an unredacted message thread reportedly indicated that the LibGen decision followed a prior escalation to “MZ,” or Mark Zuckerberg.