You’re likely aware that the AI models behind ChatGPT, Gemini, Claude, and similar chatbots are trained on vast collections of published material—hundreds of millions of books, articles, academic papers, and nearly everything else available online. Most published authors have unknowingly and without consent contributed to the development of the very AI tools that now threaten their livelihoods. That seems like it should be illegal, right? The reality is far more nuanced. “It’s very complex and there are a lot of raw feelings about what is happening, both for and against.”

Last year, in one of the first rulings of its kind, Judge William Alsup ordered Anthropic to pay a $1.5 billion copyright settlement to a group of writers whose works were used to train the company’s AI models. On the surface, this looked like a moral win for authors, but Judge Alsup actually determined that Anthropic’s AI training was lawful. What Alsup penalized the company for was pirating these books from illegal online shadow libraries. “Like any reader aspiring to be a writer, Anthropic’s LLMs trained upon works not to race ahead and replicate or supplant them - but to turn a hard corner and create something different,” the judge wrote, likening the way an LLM ingests trillions of words to a writer’s study of literature. Gellis believes the ruling is more favorable to AI companies. What’s a $1.5 billion fine to a company projecting roughly $200 billion in annual revenue by 2028? “I think it is generally good news for AI training that he looked at what was going on and really sort of thought it analogous to reading a copyrighted work as opposed to copying a copyrighted work,” Gellis said. “Copyright law hinges on copying, but it doesn’t hinge on using the work or experiencing the work, consuming the work, reading the work.”

Copyright law hasn’t been updated since 1976, which means judges are left to interpret guidelines from five decades ago when confronting legal questions that could shape the future of the AI industry. “They know that the AI model has been trained on so much stuff, and the law has not really caught up to that question.”

These cases often hinge on fair use law—specifically, whether using a copyrighted work is “transformative” enough to be legally permissible. Fair use is an exception in copyright law that allows for the use of copyrighted materials without explicit permission, protecting activities like criticism, parody, education, and other forms of commentary and iteration. Judges weigh several factors when deciding if something qualifies as fair use, including the purpose and nature of the work, the amount used, and its impact on the market. “Copyright is always about protecting and growing the market,” Henderson noted. “The courts are kind of all over the place in their reasoning [in AI cases]. What’s tending to win is if what you’re doing is you’re training on somebody’s property because your purpose is to directly compete, then the courts will frown on it… If what you’re doing is not going to compete, then the courts are tending to find ways that it will be okay.”

Henderson is referencing a case where the media and technology company Thomson Reuters sued research firm Ross Intelligence for copying its content to build a competing, AI-based legal platform. “Ross’s use is not transformative because it does not have a ‘further purpose or different character’ than Thomson Reuters’s,” Judge Stephanos Bibas wrote last year. In that instance, Judge Bibas decided that training on Reuters’ content to create a platform that would directly compete with it was not fair use. While authors could potentially argue that chatbots are competing with them by using their works to generate new, synthetic books, that argument hasn’t yet prevailed in court. When it comes to the relationship between AI and copyright, Gellis finds it useful to clarify what we’re actually discussing—the way we think about copyright in terms of AI training is quite different from how we think about copyrighting AI-generated content. In one case, Thaler v. Perlmutter, the court ruled that if a work is 100% AI-generated, it’s not copyrightable, which opens a whole new set of questions—how can we definitively prove whether a work was generated using AI, and if so, how do we know what percentage was created or assisted by AI? “If you write your novel in [Microsoft] Word and run spell check, we kind of feel comfortable with the idea of saying that Word does not own your novel,” Gellis said. “[AI] is forcing us to look at a whole bunch of decisions that we kind of ignored for a while.”

Most AI companies are still entangled in pending litigation over these issues, which means we won’t have a definitive resolution any time soon. “What you are seeing is that the initial opening volleys are being influential, and that influence itself could be undone if other courts decide different things, and it’ll take later states of litigation to figure out which one will prevail,” Gellis said. “But in the meantime, all these decisions are shaping everything that’s happening. It would be kind of foolish for the AI companies to ignore them.”