Click here to download a pdf of the full paper.
Introduction
Throughout each of the previous industrial revolutions, innovators and early adopters were able to create new industries, topple incumbents, and lay the groundwork for other innovative products and services, often termed “spillovers.” This was particularly true of general-purpose technologies (GPTs), which are processes or products that themselves enable the creation of entirely new technologies.
Artificial intelligence (AI) has the makings of a GPT, given its general application in problem solving, and could usher in a new industrial revolution. Breakthroughs have included applications related to novel medical technologies, autonomous vehicles and robotics, and advanced military targeting systems, as well as consumer-facing applications, such as agentic search, personalized tutors, and creative tools. Advancements in AI have been driven by progress in the use of machine learning (ML), a specific area of AI that develops software that enables machines to learn recursively without human instruction.
Because of AI’s immense potential, researchers, firms, and nation-states are jockeying to be leaders in the technology. This includes creating government bodies to study the potential effects of the technology, investing in domestic capabilities to manufacture key inputs to the AI stack, and attracting private investment and talent to bolster the domestic AI industry. As a GPT, AI could be diffused into a host of existing processes, while also creating new products, services, and markets. This has significant implications for national soft power, such as the strength of a nation’s economy, citizens' relationship to and embrace of technology, and impact on culture, both at home and abroad. There are also implications for hard power, including military might, statecraft, national sovereignty, and geopolitical competition.
In terms of geopolitical competition, the most significant competition is taking place between the United States and China. While the United States has been a global power since the twentieth century and the preeminent power of the twenty-first century so far, leaders of the Chinese Communist Party (CCP) have made clear their desire to change the status quo, and AI features heavily in such plans. In 2018, President Xi Jinping stated that China must “ensure that our country marches in the front ranks where it comes to theoretical research in this important area of AI, and occupies the high ground in critical and AI core technologies.” In 2023, Xi directed the People’s Liberation Army to “raise the presence of combat forces in new domains and of new qualities.”
Considering this competition with China, there is one area of law that could deal a serious blow to U.S. efforts: copyright. Nearly all the leading AI model developers in the U.S. are facing lawsuits from copyright-holders claiming that the inclusion of copyrighted material in training materials for AI models constitutes infringement. There are currently 38 open copyright cases against model developers, with plaintiffs seeking damages, asking for data sets to be destroyed and models to be retrained. If plaintiffs are successful, they would do irreparable harm to America’s AI sector, with grave implications for America’s economy, military, and geopolitical standing.
If the United States is not the world leader in AI, China will be, which would have profoundly negative consequences domestically and globally. The CCP’s expansive surveillance capabilities, which cover both the physical and digital world, are trained to snuff out any dissent or opposition to the party. PRC-based firms are already producing innovative products combining AI models with expansive amounts of data to improve sensors, radar systems, and computer vision. These are all examples of dual-use technologies, which can be useful for consumer purposes as well as military and intelligence gathering. In its own planning documents, the CCP emphasizes the whole-of-nation approach to building intelligent systems and models within China to ensure domestic firms can flourish at home and expand their dominance abroad.
With its established lead, the American AI industry can provide a superior alternative, if we allow it to evolve. Unlike in China, American AI developers will not have to abide by speech codes or socialist values to train their models, and they should be free to choose whether or not to partner with the state. The ability to train on publicly available information and the diversity of information distributed across the open internet is a distinct advantage for American AI developers that must be protected. This can allow American developers to lay the groundwork for the technology to reflect and further American values such as freedom of expression, diversity of thought, and economic dynamism if we can avoid the weaponization of copyright law.
This paper will discuss copyright law’s impact on the future of American AI development and the implications for American soft and hard power, and offer policy proposals to promote continued AI innovation in the U.S. while mitigating concerns raised by rights-holders.
