Historically, legislation and court decisions have always lagged behind technological change. But the gap between the two is even starker in the world of generative AI. The large language model (LLM) that kicked things off, ChatGPT, was only released in November 2022; the pace of development since then has been dizzying. Meanwhile, nearly 200 lawsuits have been brought against AI companies, mostly claiming copyright infringement. Very few have been resolved so far, leaving this whole area in a kind of legal limbo.
One particularly important issue is whether it is fair use to use copyright material to train AI (at least in jurisdictions where fair use rights exist). The Electronic Frontier Foundation (EFF) has a long history of defending fair use, and of trying to thwart attempts to expand copyright protections – a theme that also runs through Walled Culture the book (free digital versions available). As a recent EFF post explains, companies are once again attempting to expand the reach of copyright’s legal protection because of developments in the AI world:
Copyright owners claim that unless courts abandon 300-year-old copyright principles—and give rightsholders the power to control non-infringing works created by others—an imagined flood of AI-generated works will devastate creative markets. Under this “market dilution” theory, building generative AI tools cannot be fair use because those tools might be [sic] encourage the proliferation of competing works.
As the EFF rightly notes, if the courts took this approach, it would have serious knock-on effects:
The “market dilution” theory would eviscerate not only the fair use doctrine, but also other limits on copyright that work specifically to prevent rightsholders from unfairly suppressing competition by claiming broad ownership over tropes, genres, styles, and so on. In other words, publishers would wield unchecked veto power over any expression that might conceivably compete with a work they own.
Support for the EFF’s view that fair use applies to the training of LLMs comes from a surprising quarter: the US Department of Justice, which has just issued what it calls a “Statement of Interest of the United States” in the case brought by The New York Times against ChatGPT’s creator, OpenAI, and Microsoft, alleging copyright infringement. The US government’s argument in support of OpenAI includes an invocation of national security. It claims that:
effective integration of AI will be critical for U.S. national security capabilities. Several years ago, the Government Accountability Office (GAO) warned that “[f]ailure to adopt and effectively integrate AI technology could hinder national security” because AI can be used for tasks such as “analyzing intelligence information,” “enhancing weapon systems (such as drones and robotic ships),” and “providing recommendations on the battlefield (such as where to target missile strikes).” Over time, it has become even more apparent that “[a]dvanced AI capabilities make our Nation stronger, but also introduce new national security considerations” requiring that “the best and most secure technology is deployed to rapidly confront any and all threats to our country.”
Copyright law is relevant here because:
Rules of law that make it significantly more difficult to develop a robust AI industry in the United States therefore threaten national security and give a competitive advantage to foreign adversaries who are not so encumbered.
Specifically, the US government is worried that the New York Times in its litigation seeks to narrow fair-use doctrine to exclude the training of LLMs. The Statement of Interest writes:
That result would be inconsistent with basic copyright law principles and severely hamper “the Progress of Science and useful Arts.” U.S. CONST. art. I, § 8, cl. 8. As discussed below, publications—including independent and start-up media, in addition to the New York Times itself—are leveraging LLMs for journalistic and other purposes. And beyond the subject matter of this litigation, LLMs are already helping researchers across fields achieve major breakthroughs. Constraining LLM development under a misunderstanding of fair use doctrine would thwart such creative and scientific progress while hindering American prosperity and economic mobility.
The document goes on to explain why requiring AI companies to license material in order to train their generative AI systems is not a good solution:
An erroneous fair use ruling would hamper competition in the market for LLMs, because only the largest technology companies might have the capital necessary to pay licensing fees. And such licensing fees would disproportionately benefit legacy media outlets due to the sheer volume of their written publications.
These are some of the things that are likely to be lost when there are licensing fees, according to the US government:
if not hindered by a strained understanding of copyright law, LLMs can and should help level the playing field between mainstream and independent publishers, for several reasons. Authors with limited resources can use LLMs to compete (by, for example, using an LLM to generate an image to accompany an article—which otherwise might require a photographer or license). And LLMs can direct users to dissenting sources that offer contrary information or perspectives. It is not in the public’s interest for the largest technology companies to have an oligopoly on LLM training due to licensing entry barriers that function primarily as large subsidies for old mainstream media companies.
Politicians in the European Union would do well to bear these issues in mind now that some are calling for collective licensing to be introduced there. Whether the court hearing The New York Times’ lawsuit against OpenAI will be influenced by the intervention of the government remains to be seen. But as time goes on, there is another reason why judges might look more favourably on fair use arguments and their equivalents in other jurisdictions. This emerges from an analysis by Andres Guadamuz on his TechnoLlama blog of another lawsuit involving alleged infringement of copyright by OpenAI. It concerns a ruling from the High Court of Delhi in India regarding the case of ANI Media v OpenAI:
Asian News International (ANI) is one of India’s largest news agencies, which supplies text and video syndication to broadcasters and publishers across South Asia. They sued OpenAI alleging copyright infringement during the training phase, as well as reproduction of its articles on the output phase. This decision is an interim order deciding ANI’s application for a temporary injunction against OpenAI, so while interesting on its own right, the trial continues.
The whole blog post is well-worth reading, but here’s the summary:
First, the jurisdiction question appears to be leaning in favour of allowing lawsuits to proceed where the effects of training are experienced. Second, this case proves once again that if you cannot produce a credible infringing output, the case will rest on the input phase, where judges seem less willing to damn an entire model for copies made during training. Third, I found the reliance on the private use exception to be an astute move in a jurisdiction that has no text and data mining exception [unlike the EU].
Guadamuz highlights an interesting aspect of the ruling:
I think we often underestimate how willing judges are to look at the actual usefulness of LLMs when ruling in these copyright cases. Judges are AI users too, and here we have the Honourable Amit Bansal, who appears to understand the technology well, and this evidently shows in the decision. I was really impressed with the clarity of the opinion, particularly in the memorisation discussion, something which I hope other judges will replicate when dealing with this question in the future.
As the use of generative AI and LLMs becomes widespread and even routine in today’s society, it seems possible that this will bring with it a certain presumption in their favour that will trump even the decades-long habit of treating copyright as something sacrosanct that must be preserved and protected at all costs.
Featured image by OpenAI.
