November 4, 2025
When pirates wear suits and badges
What Happened to Piracy? Copyright Enforcement Fades as AI Giants Rise
Big Tech’s “legal” piracy? Rich get a pass, you get fines
TLDR: AI firms are accused of training on copyrighted books and paywalled research while the government looks away, leaving authors to sue. Commenters split between “let it happen or we lose” and “rich people get to pirate,” with a few insisting the law hasn’t changed and AI isn’t breaking it.
Remember when the FBI raided servers over bootleg MP3s? Now the heat’s off the downloaders and on the data hoarders. The article says AI giants like Microsoft, Meta, and OpenAI are allegedly training their models on copyrighted books and paywalled papers—think Library Genesis—while authors sue and Washington shrugs. Cue the comments section meltdown.
The strongest vibe: competitive panic vs. moral outrage. One camp says, let AI feast on paywalled science or America loses to countries that don’t care about copyright. Another camp is furious at the “rich-can-pirate” double standard, pointing to the Aaron Swartz saga (remember him?) and years of strict enforcement for regular folks. Then there’s the cynics: copyright is just a “culture tax” for peasants, and Big Tech will always collect.
Disagreements get spicy. Some argue nothing’s changed—“AI doesn’t violate copyright,” lawsuits hit real violations, end of story. Others call out Microsoft’s plot twist: from crusading against torrents to allegedly tapping the same wells for training data. The jokes fly: “Elsevier tears power the GPUs,” “exclusive pirates club—members: billionaires,” and one troll quips that “real scientists don’t publish,” pointing at corporate secrecy and Russian research going dark.
The drama isn’t about torrents anymore; it’s about who gets to copy at scale, who pays, and who’s suddenly above the rules.
Key Points
- •The article claims U.S. copyright enforcement has shifted from criminal anti-piracy actions to civil litigation as AI rises.
- •Microsoft historically led anti-piracy efforts, funding the Business Software Alliance and supporting FBI actions against foreign piracy sites.
- •In 2011, Aaron Swartz was criminally prosecuted for downloading JSTOR content, highlighting prior strict enforcement.
- •Major AI firms (Meta, Anthropic, Microsoft, Google, xAI, OpenAI) are alleged to train models on large volumes of copyrighted works.
- •A lawsuit (Kadrey et al. v. Meta) alleges Meta used a Library Genesis mirror of pirated books to train generative AI systems.