Hawley Says Congress Must Step in to Fix AI Companies’ Mass Theft of Copyrighted Works

On Wednesday, July 16, the U.S. Senate Judiciary Committee’s Subcommittee on Crime and Counterterrorism held a hearing titled “Too Big to Prosecute?: Examining the AI Industry’s Mass Ingestion of Copyrighted Works for AI Training.” Subcommittee Chair Josh Hawley (R-MO) called generative AI companies’ use of copyrighted works to train their chatbots and other large language models (LLMs) “the largest IP theft in American history” and rejected the suggestion that the courts should determine the path forward.

Hawley opened by saying that AI companies are training their models on stolen materials and have copied enough works to fill 22 Libraries of Congress. He also noted in his introductory and subsequent remarks, using public documents from the pending case between Meta and a number of authors, that Meta employees specifically warned the company on multiple occasions that using pirated works would subject them to liability. “Meta’s conduct is not an exception, it’s the rule in the AI space,” Hawley said. “Do whatever you want and count on lawyers and lobbyists to fix it later. They care about power and money.”

Hawley added that the AI companies’ arguments that the United States needs to make technological advancement easy in order to maintain dominance in the AI space fall flat. “When they say, ‘we can’t let China beat us,’ what they’re really saying is ‘give us truckloads of cash and let us steal everything from you and make billions on it,’” Hawley said.

The hearing included testimony from five witnesses—three professors, one attorney representing the authors against Meta, and bestselling author David Baldacci. Professor Bhamati Viswanathan of New England Law School said it’s clear that what AI companies are doing is illegal, particularly in the context of using pirated works for training. “If you and I stole books from the library or from a bookstore and said, ‘I need to train, I need to learn, I need to develop my mind,’ we wouldn’t argue that’s fair use, we’d say ‘you can’t steal the materials even for a good cause,’” Viswanathan said. But the AI companies are going beyond that by “going to pirate websites [and] stealing the materials that have already been stolen—it’s a crime compounding a crime,” she added.

Bestselling author Baldacci noted that his son asked ChatGPT to write a plot that read like a David Baldacci novel and “in about five seconds three pages came up that had elements of pretty much every book I’d ever written, including plot lines, twists, character names, narrative, the works—that’s when I found out the AI community had taken most of my novels without permission and fed them into their machine learning system.”

Baldacci dismissed the argument that this process is no different than an aspiring writer reading other authors and learning from them in order to create their own original works. “I can tell you from personal experience that is flatly wrong,” he said. As an example, Baldacci explained that, when he was still an aspiring writer himself, he read everything John Irving ever wrote “and none of my novels read remotely like a John Irving novel. Unlike AI, I can’t remember every line Irving ever wrote, every detail about his characters and plots.” The AI companies’ arguments that it would be too difficult to license the works is comical, Baldacci continued. “I was once a trial lawyer; if I had made that argument in court, I would either have been laughed out of the courtroom or held in contempt by the judge, and rightly so.” He also noted that if words were all the AI companies needed, they could have simply fed every dictionary in the world into the LLMs, but what they actually sought to copy is “complete, well-crafted, living, breathing stories, with characters that seemed real, plots that made sense, dialogue that appeared genuine–humanity on the page. In sum, they needed us.”

Professor Edward Lee of Santa Clara University School of Law defended the training of AI models with copyrighted works, arguing it is an imperative for the Trump Administration’s national priority of maintaining dominance over China. Lee quoted the unofficial U.S. “AI Czar” under Trump, David Sacks, who has said that if there is no pathway to fair use in AI training we will lose the race with China.

But Hawley pushed back, asking Lee if he thought the Committee should allow “an unelected AI Czar” to decide what the rights of American citizens are? Lee replied “no,” but said that the courts should decide, and are in the process of doing so now. “It’s not a time for Congress to intervene in deciding these difficult [issues],” Lee said.

However, Hawley rejected this approach, pointing to more emails made public in the Meta case as evidence that AI companies are engaging in willful theft. “If the answer is that the biggest corporation in the world worth trillions of dollars can come take an individual author’s work like Mr. Baldacci, lie about it, hide it, profit off of it, and there’s nothing our law does about that, we need to change the law,” said Hawley, adding to applause from attendees: “We’ve got to do something about this.”

Other witnesses who testified at the hearing included Maxwell Pritt of Boies, Schiller & Flexner, who is representing the authors in the case against Meta; and Michael Smith, Professor of Information Technology and Marketing at Carnegie Mellon University.