Meta’s AI Copyright War Has Hollywood Losing Sleep

AI is the Wild West right now, and the lawmen are coming for it. Every day, we're tracking new laws and lawsuits because we're getting landmark cases left and right.

And there will be staggering effects in Hollywood as these cases settle, since AI has its grip on pretty much every aspect right now.

The biggest suit in progress with Meta has huge implications for where movies and TV go with AI in the future.

According to a recent report from Los Angeles Magazine , Meta’s ongoing legal battle over AI training data has entered a new phase. It’s a class-action lawsuit that names Mark Zuckerberg personally and targets the very foundation of how generative AI is built and trained.

Let's dive in.

Trained on Stolen Goods

Okay, so the heart of this new lawsuit, which was brought by a coalition of major publishers (including Hachette and Macmillan) and bestselling author Scott Turow, alleges that Meta actively sought out and torrented millions of copyrighted works to train its models.

The TL:DR is that they allegedly stole a bunch of stuff and then used it to train AI.

Now, imagine what they did with movies and TV shows... which are torrented all the time and traded on the web.

We’ve all spent the last year hearing about "fair use" and "transformative" training of generative AI. But if the source material was acquired through stealing, that defense crumbles.

And they owe filmmakers a lot of money.

Credit: Meta

Why Zuckerberg is in the Crosshairs

Usually, CEOs are shielded in these kinds of lawsuits, but the plaintiffs claim that Mark Zuckerberg "personally authorized and actively encouraged" the use of these pirated collections to catch up in the AI arms race.

That would not be good for him, and it would set a precedent that if a CEO can be held liable for the data-gathering practices of their AI, you can bet every studio head in Hollywood will be on the phone with general counsel to double-check their own AI initiatives to make sure it's all above board.

The "Infinite Substitution Machine"

This is the term that should keep every filmmaker awake at night. The lawsuit argues that Meta’s AI isn't just "learning" from books; it is being designed as an "infinite substitution machine."

The idea is that Meta's AI program, called Llama, can recreate an author’s specific style and voice so accurately that it creates a market substitute for the original work.

To put this in Hollywood terms, imagine a tool that doesn't just "write a script," but writes a script exactly like Aaron Sorkin or Greta Gerwig, using their proprietary "voice" as the engine.

What's at stake in a court deciding whether you own your own "voice" or if AI is allowed to mimic it in ways that will be used to replace you.

Why Hollywood is Watching

If publishers can’t protect their books from being fed into Llama, there is nothing stopping AI companies from doing the same with every screenplay online and every movie and TV show they can get their hands on as well.

Hollywood runs on personal relationships and reputations. Seeing a tech titan personally sued for "unmitigated piracy" changes the optics of AI from "inevitable innovation" to "potential legal liability."

We have to hope that these courts uphold our own legal holdings.

Summing It All Up

This is a tricky situation. We're seeing pirated works used to train AI, and we're trying to figure out who had knowledge of it, and also protect our voices from future piracy as well.

This is an ongoing lawsuit we'll keep an eye on.

Let us know in the comments.