AI Absorbs Millions of Books Without Permission: The Battle Between AI and Copyright, A Future Shaken in Court

AI's "Book Plundering"—The Battle Between Creators, Courts, and Congress

AI navigating the seas of copyright—it is becoming more than just a creative partner, now seen as a potential threat. According to The Washington Post, major AI companies (OpenAI, Microsoft, Anthropic, etc.) have allegedly collected millions of books as training data without authors' consent, potentially infringing on copyrights. Authors have condemned this as "the largest theft of intellectual property in American history" and have begun to fight back through lawsuits The Washington Post.

Authors' Anger: What Lies Beyond the Lawsuits

Numerous authors, including legal thriller writer David Baldacci, have filed lawsuits against OpenAI, Microsoft, and Anthropic. They are concerned that AI products could automatically generate all kinds of derivative works, potentially devaluing their creations and harming their income The Washington PostThe Week.

Some authors are also focusing on the fact that Anthropic allegedly downloaded up to 7 million pirated books from "shadow libraries" like LibGen and PiLiMi. Judge William Alsup of the San Francisco District Court acknowledged the "transformative" nature of the training but ruled that the mass possession of pirated copies for storage is illegal, allowing the class-action lawsuit to proceed .

Court Decisions: How Far Does Fair Use Go?

As similar rulings continue against Meta and Anthropic, some cases have emerged where fair use claims are supported. In June 2025, Judges Alsup (Anthropic) and Chhabria (Meta) acknowledged the collection of books but objected to the stockpiling of pirated copies, ruling that lawsuits should continue regarding large-scale commercial use.

These rulings are setting precedents for the legal framework of "AI book usage." Future disputes are expected to focus on court records, the "creativity" of AI outputs, and damages to affected authors

Congress Reacts: The Train Act and Potential Legal Reforms

Some senators are proposing the "Train Act," which would allow authors to inquire whether their works are included in AI training data, aiming to enhance transparency through legal reforms

Social Media Reactions: "Anger," "Irony," and "Calls to Action"

On Twitter and Reddit, authors' anger and the literary world's sense of crisis have erupted. Comments like "It's real funny that AI is 'stealing my book,'" "How am I supposed to make a living without copyright?" and "Act before creativity becomes 'poisoned'" have been circulating. Meanwhile, fair use supporters argue, "Acquisition and 'use' of books are separate issues," and "If transformative, secondary use is okay," engaging in rational discussions

■ Summary: The Framework of Conflict and Future Directions