I. Introduction

We thank the National Telecommunications and Information Administration (NTIA) for the opportunity to contribute to this request for comments (RFC) in the “Dual Use Foundation Artificial Intelligence Models with Widely Available Model Weights” proceeding. In these comments, we endeavor to offer recommendations to foster the innovative and responsible production of artificial intelligence (AI), encompassing both open-source and proprietary models. Our comments are guided by a belief in the transformative potential of AI, while recognizing NTIA’s critical role in guiding the development of regulations that not only protect consumers but also enable this dynamic field to flourish. The agency should seek to champion a balanced and forward-looking approach toward AI technologies that allows them to evolve in ways that maximize their social benefits, while navigating the complexities and challenges inherent in their deployment.

NTIA’s question “How should [the] potentially competing interests of innovation, competition, and security be addressed or balanced?”[1] gets to the heart of ongoing debates about AI regulation. There is no panacea to be discovered, as all regulatory choices require balancing tradeoffs. It is crucial to bear this in mind when evaluating, e.g., regulatory proposals that implicitly treat AI as inherently dangerous and regard as obvious that stringent regulation is the only effective strategy to mitigate such risks.[2] Such presumptions discount AI’s unknown but potentially enormous capacity to produce innovation, and inadequately account for other tradeoffs inherent to imposing a risk-based framework (e.g., requiring disclosure of trade secrets or particular kinds of transparency that could yield new cybersecurity attack vectors). Adopting an overly cautious stance risks not only stifling AI’s evolution, but may also preclude a fulsome exploration of its potential to foster social, economic, and technological advancement. A more restrictive regulatory environment may also render AI technologies more homogenous and smother development of the kinds of diverse AI applications needed to foster robust competition and innovation.

We observe this problematic framing in the executive order (EO) that serves as the provenance of this RFC.[3] The EO repeatedly proclaims the importance of “[t]he responsible development and use of AI” in order to “mitigate[e] its substantial risks.”[4] Specifically, the order highlights concerns over “dual-use foundation models”—i.e., AI systems that, while beneficial, could pose serious risks to national security, national economic security, national public health, or public safety.[5] Concerningly, one of the categories the EO flags as illicit “dual use” are systems “permitting the evasion of human control or oversight through means of deception or obfuscation.”[6] This open-ended category could be interpreted so broadly that essentially any general-purpose generative-AI system would classify.

The EO also repeatedly distinguishes “open” versus “closed” approaches to AI development, while calling for “responsible” innovation and competition.[7] On our reading, the emphasis the EO places on this distinction raises alarm bells about the administration’s inclination to stifle innovation through overly prescriptive regulatory frameworks, diminishment of the intellectual property rights that offer incentives for innovation, and regulatory capture that favors incumbents over new entrants. In favoring one model of AI development over another, the EO’s prescriptions could inadvertently hamper the dynamic competitive processes that are crucial both for technological progress and for the discovery of solutions to the challenges that AI technology poses.

Given the inchoate nature of AI technology—much less the uncertain markets in which that technology will ultimately be deployed and commercialized—NTIA has an important role to play in elucidating for policymakers the nuances that might lead innovators to choose an open or closed development model, without presuming that one model is inherently better than the other—or that either is necessarily “dangerous.” Ultimately, the preponderance of AI risks will almost certainly emerge idiosyncratically. It will be incumbent on policymakers to address such risks in an iterative fashion as they become apparent. For now, it is critical to resist the urge to enshrine crude and blunt categories for the heterogeneous suite of technologies currently gathered under the broad banner of  “AI.”

Section II of these comments highlights the importance of grounding AI regulation in actual harms, rather than speculative risks, while outlining the diversity of existing AI technologies and the need for tailored approaches. Section III starts with discussion of some of the benefits and challenges posed by both open and closed approaches to AI development, while cautioning against overly prescriptive definitions of “openness” and advocating flexibility in regulatory frameworks. It proceeds to examine the EO’s prescription to regulate so-called “dual-use” foundation models, underscoring some potential unintended consequences for open-source AI development and international collaboration. Section IV offers some principles to craft an effective regulatory model for AI, including distinguishing between low-risk and high-risk applications, avoiding static regulatory approaches, and adopting adaptive mechanisms like regulatory sandboxes and iterative rulemaking. Section V concludes.

II. Risk Versus Harm in AI Regulation

In many of the debates surrounding AI regulation, disproportionate focus is placed on the need to mitigate risks, without sufficient consideration of the immense benefits that AI technologies could yield. Moreover, because these putative risks remain largely hypothetical, proposals to regulate AI descend quickly into an exercise in shadowboxing.

Indeed, there is no single coherent definition of what even constitutes “AI.” The term encompasses a wide array of technologies, methodologies, and applications, each with distinct characteristics, capabilities, and implications for society. From foundational models that can generate human-like text, to algorithms capable of diagnosing diseases with greater accuracy than human doctors, to “simple” algorithms that facilitate a more tailored online experience, AI applications and their underlying technologies are as varied as they are transformative.

This diversity has profound implications for the regulation and development of AI. Very different regulatory considerations are relevant to AI systems designed for autonomous vehicles than for those used in financial algorithms or creative-content generation. Each application domain comes with its own set of risks, benefits, ethical dilemmas, and potential social impacts, necessitating tailored approaches to each use case. And none of these properties of AI map clearly onto the “open” and “closed” designations highlighted by the EO and this RFC. This counsels for focus on specific domains and specific harms, rather than how such technologies are developed.[8]

As in prior episodes of fast-evolving technologies, what is considered cutting-edge AI today may be obsolete tomorrow. This rapid pace of innovation further complicates the task of crafting policies and regulations that will be both effective and enduring. Policymakers and regulators must navigate this terrain with a nuanced understanding of AI’s multifaceted nature, including by embracing flexible and adaptive regulatory frameworks that can accommodate AI’s continuing evolution.[9] A one-size-fits-all approach could inadvertently stifle innovation or entrench the dominance of a few large players by imposing barriers that disproportionately affect smaller entities or emerging technologies.

Experts in law and economics have long scrutinized both market conduct and regulatory rent seeking that serve to enhance or consolidate market power by disadvantaging competitors, particularly through increasing the costs incurred by rivals.[10] Various tactics may be employed to undermine competitors or exclude them from the market that do not involve direct price competition. It is widely recognized that “engaging with legislative bodies or regulatory authorities to enact regulations that negatively impact competitors” produces analogous outcomes.[11] It is therefore critical that the emerging markets for AI technologies not engender opportunities for firms to acquire regulatory leverage over rivals. Instead, recognizing the plurality of AI technologies and encouraging a multitude of approaches to AI development could help to cultivate a more vibrant and competitive ecosystem, driving technological progress forward and maximizing AI’s potential social benefits.

This overarching approach counsels skepticism about risk-based regulatory frameworks that fail to acknowledge how the theoretical harms of one type of AI system may be entirely different from those of another. Obviously, the regulation of autonomous drones is a very different sort of problem than the regulation of predictive policing or automated homework tutors. Even within a single circumscribed domain of generative AI—such as “smart chatbots” like ChatGPT or Claude—different applications may present entirely different kinds of challenges. A highly purpose-built version of such a system might be employed by government researchers to develop new materiel for the U.S. Armed Forces, while a general-purpose commercial chatbot would employ layers of protection to ensure that ordinary users couldn’t learn how to make advanced weaponry. Rather treating “chatbots” as possible vectors for weapons development, a more appropriate focus would target high-capability systems designed to assist in developing such systems. Were it the case that a general-purpose chatbot inadvertently revealed some information on building weapons, all incentives would direct that AI’s creators to treat that as a bug to fix, not a feature to expand.

Take, for example, the recent public response to the much less problematic AI-system malfunctions that accompanied Google’s release of its Gemini program.[12] Gemini was found to generate historically inaccurate images, such as ethnically diverse U.S. senators from the 1800s, including women.[13] Google quickly acknowledged that it did not intend for Gemini to create inaccurate historical images and turned off the image-generation feature to allow time for the company to work on significant improvements before re-enabling it.[14] While Google blundered in its initial release, it had every incentive to discover and remedy the problem. The market response provided further incentive for Google to get it right in the future.[15] Placing the development of such systems under regulatory scrutiny because some users might be able to jailbreak a model and generate some undesirable material would create disincentives to the production of AI systems more generally, with little gained in terms of public safety.

Rather than focus on the speculative risks of AI, it is essential to ground regulation in the need to address tangible harms that stem from the observed impacts of AI technologies on society. Moreover, focusing on realistic harms would facilitate a more dynamic and responsive regulatory approach. As AI technologies evolve and new applications emerge, so too will the  potential harms. A regulatory framework that prioritizes actual harms can adapt more readily to these changes, enabling regulators to update or modify policies in response to new evidence or social impacts. This flexibility is particularly important for a field like AI, where technological advancements could quickly outpace regulation, creating gaps in oversight that may leave individuals and communities vulnerable to harm.

Furthermore, like any other body of regulatory law, AI regulation must be grounded in empirical evidence and data-driven decision making. Demanding a solid evidentiary basis as a threshold for intervention would help policymakers to avoid the pitfalls of reacting to sensationalized or unfounded AI fears. This would not only enhance regulators’ credibility with stakeholders, but would also ensure that resources are dedicated to addressing the most pressing and substantial issues arising from the development of AI.

III. The Regulation of Foundation Models

NTIA is right to highlight the tremendous promise that attends the open development of AI technologies:

Dual use foundation models with widely available weights (referred to here as open foundation models) could play a key role in fostering growth among less resourced actors, helping to widely share access to AI’s benefits…. Open foundation models can be readily adapted and fine-tuned to specific tasks and possibly make it easier for system developers to scrutinize the role foundation models play in larger AI systems, which is important for rights- and safety-impacting AI systems (e.g. healthcare, education, housing, criminal justice, online platforms etc.)

…Historically, widely available programming libraries have given researchers the ability to simultaneously run and understand algorithms created by other programmers. Researchers and journals have supported the movement towards open science, which includes sharing research artifacts like the data and code required to reproduce results.[16]

The RFC proceeds to seek input on how to define “open” and “widely available.”[17] These, however, are the wrong questions. NTIA should instead proceed from the assumption that there are no harms inherent to either “open” or “closed” development models; it should be seeking input on anything that might give rise to discrete harms in either open or closed systems.

NTIA can play a valuable role by recommending useful alterations to existing law where gaps currently exist, regardless of the business or distribution model employed by the AI developer. In short, there is nothing necessarily more or less harmful about adopting an “open” or a “closed” approach to software systems. The decision to pursue one path over the other will be made based on the relevant tradeoffs that particular firms face. Embedding such distinctions in regulation is arbitrary, at best, and counterproductive to the fruitful development of AI, at worst.

A. ‘Open’ or ‘Widely Available’ Model Weights

To the extent that NTIA is committed to drawing distinctions between “open” and “closed” approaches to developing foundation models, it should avoid overly prescriptive definitions of what constitutes “open” or “widely available” model weights that could significantly hamper the progress and utility of AI technologies.

Imposing narrow definitions risks creating artificial boundaries that fail to accurately reflect AI’s technical and operational realities. They could also inadvertently exclude or marginalize innovative AI models that fall outside those rigid parameters, despite their potential to contribute positively to technological advancement and social well-being. For instance, a definition of “open” that requires complete public accessibility without any form of control or restriction might discourage organizations from sharing their models, fearing misuse or loss of intellectual property.

Moreover, prescriptive definitions could stifle the organic growth and evolution of AI technologies. The AI field is characterized by its rapid pace of change, where today’s cutting-edge models may become tomorrow’s basic tools. Prescribing fixed criteria for what constitutes “openness” or “widely available” risks anchoring the regulatory landscape to this specific moment in time, leaving the regulatory framework less able to adapt to future developments and innovations.

Given AI developers’ vast array of applications, methodologies, and goals, it is imperative that any definitions of “open” or “widely available” model weights embrace flexibility. A flexible approach would acknowledge how the various stakeholders within the AI ecosystem have differing needs, resources, and objectives, from individual developers and academic researchers to startups and large enterprises. A one-size-fits-all definition of “openness” would fail to accommodate this diversity, potentially privileging certain forms of innovation over others and skewing the development of AI technologies in ways that may not align with broader social needs.

Moreover, flexibility in defining “open” and “widely available” must allow for nuanced understandings of accessibility and control. There can, for example, be legitimate reasons to limit openness, such as protecting sensitive data, ensuring security, and respecting intellectual-property rights, while still promoting a culture of collaboration and knowledge sharing. A flexible regulatory approach would seek a balanced ecosystem where the benefits of open AI models are maximized, and potential risks are managed effectively.

B. The Benefits of ‘Open’ vs ‘Closed’ Business Models

NTIA asks:

What benefits do open model weights offer for competition and innovation, both in the AI marketplace and in other areas of the economy? In what ways can open dual-use foundation models enable or enhance scientific research, as well as education/training in computer science and related fields?[18]

An open approach to AI development has obvious benefits, as NTIA has itself acknowledged in other contexts.[19] Open-foundation AI models represent a transformative force, characterized by their accessibility, adaptability, and potential for widespread application across various sectors. The openness of these models may serve to foster an environment conducive to innovation, wherein developers, researchers, and entrepreneurs can build on existing technologies to create novel solutions tailored to diverse needs and challenges.

The inherent flexibility of open-foundation models can also catalyze a competitive market, encouraging a healthy ecosystem where entities ranging from startups to established corporations may all participate on roughly equal footing. By lowering some entry barriers related to access to basic AI technologies, this competitive environment can further drive technological advancements and price efficiencies, ultimately benefiting consumers and society at-large.

But more “closed” approaches can also prove very valuable. As NTIA notes in this RFC, it is rarely the case that a firm pursues a purely open or closed approach. These terms exist along a continuum, and firms blend models as necessary.[20] And just as firms readily mix elements of open and closed business models, a regulator should be agnostic about the precise mix that firms employ, which ultimately must align with the realities of market dynamics and consumer preferences.

Both open and closed approaches offer distinct benefits and potential challenges. For instance, open approaches might excel in fostering a broad and diverse ecosystem of applications, thereby appealing to users and developers who value customization and variety. They can also facilitate a more rapid dissemination of innovation, as they typically impose fewer restrictions on the development and distribution of new applications. Conversely, closed approaches, with their curated ecosystems, often provide enhanced security, privacy, and a more streamlined user experience. This can be particularly attractive to users less inclined to navigate the complexities of open systems. Under the right conditions, closed systems can likewise foster a healthy ecosystem of complementary products.

The experience of modern digital platforms demonstrates that there is no universally optimal approach to structuring business activities, thus illustrating the tradeoffs inherent in choosing among open and closed business models. The optimal choice depends on the specific needs and preferences of the relevant market participants. As Jonathan M. Barnett has noted:

Open systems may yield no net social gain over closed systems, can pose a net social loss under certain circumstances, and . . . can impose a net social gain under yet other circumstances.^[21]

Similar considerations apply in the realm of AI development. Closed or semi-closed ecosystems can offer such advantages as enhanced security and curated offerings, which may appeal to certain users and developers. These benefits, however, may come at the cost of potentially limited innovation, as a firm must rely on its own internal processes for research and development. Open models, on the other hand, while fostering greater collaboration and creativity, may also introduce risks related to quality control, intellectual-property protection, and a host of other concerns that may be better controlled in a closed business model. Even along innovation dimensions, closed platforms can in many cases outperform open models.

With respect to digital platforms like the App Store and Google Play Store, there is a “fundamental welfare tradeoff between two-sided proprietary…platforms and two-sided platforms which allow ‘free entry’ on both sides of the market.”^[22] Consequently, “it is by no means obvious which type of platform will create higher product variety, consumer adoption and total social welfare.”^[23]

To take another example, consider the persistently low adoption rates for consumer versions of the open-source Linux operating system, versus more popular alternatives like Windows or MacOS.[24] A closed model like Apple’s MacOS is able to outcompete open solutions by better leveraging network effects and developing a close relationship with end users.[25] Even in this example, adoption of open versus closed models varies across user types, with, e.g., developers showing a strong preference for Linux over Mac, and only a slight preference for Windows over Linux.[26] This underscores the point that the suitability of an open or closed model varies not only by firm and product, nor even solely by user, but by the unique fit of a particular model for a particular user in a particular context. Many of those Linux-using developers will likely not use it on their home computing device, for example, even if they prefer it for work.

The dynamics among consumers and developers further complicate prevailing preferences for open or closed models. For some users, the security and quality assurance provided by closed ecosystems outweigh the benefits of open systems’ flexibility. On the developer side, the lower barriers to entry in more controlled ecosystems that smooth the transaction costs associated with developing and marketing applications can democratize application development, potentially leading to greater innovation within those ecosystems. Moreover, distinctions between open and closed models can play a critical role in shaping inter-brand competition. A regulator placing its thumb on the business-model scale would push the relevant markets toward less choice and lower overall welfare.[27]

By differentiating themselves through a focus on ease-of-use, quality, security, and user experience, closed systems contribute to a vibrant competitive landscape where consumers have clear choices between differing “brands” of AI. Forcing an AI developer to adopt practices that align with a regulator’s preconceptions about the relative value of “open” and “closed” risks homogenizing the market and diminishing the very competition that spurs innovation and consumer choice.

Consider some of the practical benefits sought by deployers when choosing between open and closed models. For example, it’s not straightforward to say close is inherently better than open when considering issues of data sharing or security; even here, there are tradeoffs. Open innovation in AI—characterized by the sharing of data, algorithms, and methodologies within the research community and beyond—can mitigate many of the risks associated with model development. This openness fosters a culture of transparency and accountability, where AI models and their applications are subject to scrutiny by a broad community of experts, practitioners, and the general public. This collective oversight can help to identify and address potential safety and security concerns early in the development process, thus enhancing AI technologies’ overall trustworthiness.

By contrast, a closed system may implement and enforce standardized security protocols more quickly. A closed system may have a sharper, more centralized focus on providing data security to users, which may perform better along some dimensions. And while the availability of code may provide security in some contexts, in other circumstances, closed systems perform better.[28]

In considering ethical AI development, different types of firms should be free to experiment with different approaches, even blending them where appropriate. For example, Claude’s approach to “Collective Constitutional AI” adopts what is arguably a “semi-open” model, blending proprietary elements with certain aspects of openness to foster innovation, while also maintaining a level of control.[29] This model might strike an appropriate balance, in that it ensures some degree of proprietary innovation and competitive advantage while still benefiting from community feedback and collaboration.

On the other hand, fully open-source development could lead to a different, potentially superior result that meets a broader set of needs through community-driven evolution and iteration. There is no way to determine, ex ante, that either an open or a closed approach to AI development will inherently provide superior results for developing “ethical” AI. Each has its place, and, most likely, the optimal solutions will involve elements of both approaches.

In essence, codifying a regulatory preference for one business model over the other would oversimplify the intricate balance of tradeoffs inherent to platform ecosystems. Economic theory and empirical evidence suggest that both open and closed platforms can drive innovation, serve consumer interests, and stimulate healthy competition, with all of these considerations depending heavily on context. Regulators should therefore aim for flexible policies that support coexistence of diverse business models, fostering an environment where innovation can thrive across the continuum of openness.

C. Dual-Use Foundation Models and Transparency Requirements

The EO and the RFC both focus extensively on so-called “dual-use” foundation models:

Foundation models are typically defined as, “powerful models that can be fine-tuned and used for multiple purposes.” Under the Executive Order, a “dual-use foundation model” is “an AI model that is trained on broad data; generally uses self-supervision, contains at least tens of billions of parameters; is applicable across a wide range of contexts; and that exhibits, or could be easily modified to exhibit, high levels of performance at tasks that pose a serious risk to security, national economic security, national public health or safety, or any combination of those matters….”[30]

But this framing will likely do more harm than good. As noted above, the terms “AI” or “AI model” are frequently invoked to refer to very different types of systems. Further defining these models as “dual use” is also unhelpful, as virtually any tool in existence can be “dual use” in this sense. Certainly, from a certain perspective, all software—particularly highly automated software—can pose a serious risk to “national security” or “safety.” Encryption and other privacy-protecting tools certainly fit this definition.[31] While it is crucial to mitigate harms associated with the misuse of AI technologies, the blanket treatment of all foundation models under this category is overly simplistic.

The EO identifies certain clear risks, such as the possibility that models could aid in the creation of chemical, biological, or nuclear weaponry. These categories are obvious subjects for regulatory control, but the EO then appears to open a giant definitional loophole that threatens to subsume virtually any useful AI system. It employs expansive terminology to describe a more generalized threat—specifically, that dual-use models could “[permit] the evasion of human control or oversight through means of deception or obfuscation.”[32] Such language could encompass a wide array of general-purpose AI models. Furthermore, by labeling systems capable of bypassing human decision making as “dual use,” the order implicitly suggests that all AI could pose such risk as warrants national-security levels of scrutiny.

Given the EO’s broad definition of AI as “a machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations, or decisions influencing real or virtual environments,” numerous software systems not typically even considered AI might be categorized as “dual-use” models.[33] Essentially, any sufficiently sophisticated statistical-analysis tool could qualify under this definition.

A significant repercussion of the EO’s very broad reporting mandates for dual-use systems, and one directly relevant to the RFC’s interest in promoting openness, is that these might chill open-source AI development.[34] Firms dabbling in AI technologies—many of which might not consider their projects to be dual use—might keep their initiatives secret until they are significantly advanced. Faced with the financial burden of adhering to the EO’s reporting obligations, companies that lack a sufficiently robust revenue model to cover both development costs and legal compliance might be motivated to dodge regulatory scrutiny in the initial phases, consequently dampening the prospects for transparency.

It is hard to imagine how open-source AI projects could survive in such an environment. Open-source AI code libraries like TensorFlow[35] and PyTorch[36] foster remarkable innovation by allowing developers to create new applications that use cutting-edge models. How could a paradigmatic startup developer working out of a garage genuinely commit to open-source development if tools like these fall under the EO’s jurisdiction? Restricting access to the weights that models use—let alone avoiding open-source development entirely—may hinder independent researchers’ ability to advance the forefront of AI technology.

Moreover, scientific endeavors typically benefit from the contributions of researchers worldwide, as collaborative efforts on a global scale are known to fast-track innovation. The pressure the EO applies to open-source development of AI tools could curtail international cooperation, thereby distancing American researchers from crucial insights and collaborations. For example, AI’s capacity to propel progress in numerous scientific areas is potentially vast—e.g., utilizing MRI images and deep learning for brain-tumor diagnoses[37] or employing machine learning to push the boundaries of materials science.[38] Such research does not benefit from stringent secrecy, but thrives on collaborative development. Enabling a broader community to contribute to and expand upon AI advancements supports this process.

Individuals respond to incentives. Just as how well-intentioned seatbelt laws paradoxically led to an uptick in risky driving behaviors,[39] ill-considered obligations placed on open-source AI developers could unintentionally stifle the exchange of innovative concepts crucial to maintain the United States’ leadership in AI innovation.

IV. Regulatory Models that Support Innovation While Managing Risks Effectively

In the rapidly evolving landscape of artificial intelligence (AI), it is paramount to establish governance and regulatory frameworks that both encourage innovation and ensure safety and ethical integrity. An effective regulatory model for AI should be adaptive, principles-based, and foster a collaborative environment among regulators, developers, researchers, and the broader community. A number of principles can help in developing this regime.

A. Low-Risk vs High-Risk AI

First, a clear distinction should be made between low-risk AI applications that enhance operational efficiency or consumer experience and high-risk applications that could have significant safety implications. Low-risk applications like search algorithms and chatbots should be governed by a set of baseline ethical guidelines and best practices that encourage innovation, while ensuring basic standards are met. On the other hand, high-risk applications—such as those used by law enforcement or the military—would require more stringent review processes, including impact assessments, ethical reviews, and ongoing monitoring to mitigate potentially adverse effects.

Contrast this with the recently enacted AI Act in the European Union, and its decision to create presumptions of risk for general purpose AI (GPAI) systems, such as large language models (LLMs), that present what the EU has termed so-called “systemic risk.”[40] Article 3(65) of the AI Act defines systemic risk as “a risk that is specific to the high-impact capabilities of general-purpose AI models, having a significant impact on the Union market due to their reach, or due to actual or reasonably foreseeable negative effects on public health, safety, public security, fundamental rights, or the society as a whole, that can be propagated at scale across the value chain.”[41]

This definition bears similarities to the “Hand formula” in U.S. tort law, which balances the burden of precautions against the probability and severity of potential harm to determine negligence.[42] The AI Act’s notion of systemic risk, however, is applied more broadly to entire categories of AI systems based on their theoretical potential for widespread harm, rather than on a case-by-case basis.

The designation of LLMs as posing “systemic risk” is problematic for several reasons. It creates a presumption of risk merely based on a GPAI system’s scale of operations, without any consideration of the actual likelihood or severity of harm in specific use cases. This could lead to unwarranted regulatory intervention and unintended consequences that hinder the development and deployment of beneficial AI technologies. And this broad definition of systemic risk gives regulators significant leeway to intervene in how firms develop and release their AI products, potentially blocking access to cutting-edge tools for European citizens, even in the absence of tangible harms.

While it is important to address potential risks associated with AI systems, the AI Act’s approach risks stifling innovation and hindering the development of beneficial AI technologies within the EU.

B. Avoid Static Regulatory Approaches

AI regulators are charged with overseeing a dynamic and rapidly developing market, and should therefore avoid erecting a rigid framework that force new innovations into ill-fitting categories. The “regulatory sandbox” may provide a better model to balance innovation with risk management. By allowing developers to test and refine AI technologies in a controlled environment under regulatory oversight, sandboxes can be used to help identify and address potential issues before wider deployment, all while facilitating dialogue between innovators and regulators. This approach not only accelerates the development of safe and ethical AI solutions, but also builds mutual understanding and trust. Where possible, NTIA should facilitate policy experimentation with regulatory sandboxes in the AI context.

Meta’s Open Loop program is an example of this kind of experimentation.[43] This program is a policy prototyping research project focused on evaluating the National Institute of Standards and Technology (NIST) AI Risk Management Framework (RMF) 1.0.[44] The goal is to assess whether the framework is understandable, applicable, and effective in assisting companies to identify and manage risks associated with generative AI. It also provides companies an opportunity to familiarize themselves with the NIST AI RMF and its application in risk-management processes for generative AI systems. Additionally, it aims to collect data on existing practices and offer feedback to NIST, potentially influencing future RMF updates.

1. Regulation as a discovery process

Another key principle is to ensure that regulatory mechanisms are adaptive. Some examples of adaptive mechanisms are iterative rulemaking and feedback loops that allow regulations to be updated continuously in response to new developments and insights. Such mechanisms enable policymakers to respond swiftly to technological breakthroughs, ensuring that regulations remain relevant and effective, without stifling innovation.

Geoffrey Manne & Gus Hurwitz have recently proposed a framework for “regulation as a discovery process” that could be adapted to AI.[45] They argue for a view of regulation not merely as a mechanism for enforcing rules, but as a process for discovering information that can inform and improve regulatory approaches over time. This perspective is particularly pertinent to AI, where the pace of innovation and the complexity of technologies often outstrip regulators’ understanding and ability to predict future developments. This framework:

in its simplest formulation, asks regulators to consider that they might be wrong. That they might be asking the wrong questions, collecting the wrong information, analyzing it the wrong way—or even that Congress has given them the wrong authority or misunderstood the problem that Congress has tasked them to address.[46]

That is to say, an adaptive approach to regulation requires epistemic humility, with the understanding that, particularly for complex, dynamic industries:

there is no amount of information collection or analysis that is guaranteed to be “enough.” As Coase said, the problem of social cost isn’t calculating what those costs are so that we can eliminate them, but ascertaining how much of those social costs society is willing to bear.[47]

In this sense, modern regulators’ core challenge is to develop processes that allow for iterative development of knowledge, which is always in short supply. This requires a shift in how an agency conceptualizes its mission, from one of writing regulations to one of assisting lawmakers to assemble, filter, and focus on the most relevant and pressing information needed to understand a regulatory subject’s changing dynamics.[48]

As Hurwitz & Manne note, existing efforts to position some agencies as information-gathering clearinghouses suffer from a number of shortcomings—most notably, that they tend to operate on an ad hoc basis, reporting to Congress in response to particular exigencies.[49] The key to developing a “discovery process” for AI regulation would instead require setting up ongoing mechanisms to gather and report on data, as well as directing the process toward “specifications for how information should be used, or what the regulator anticipated to find in the information, prior to its collection.”[50]

Embracing regulation as a discovery process means acknowledging the limits of our collective knowledge about AI’s potential risks and benefits. This underscores why regulators should prioritize generating and utilizing new information through regulatory experiments, iterative rulemaking, and feedback loops. A more adaptive regulatory framework could respond to new developments and insights in AI technologies, thereby ensuring that regulations remain relevant and effective, without stifling innovation.

Moreover, Hurwitz & Manne highlight the importance of considering regulation as an information-producing activity.[51] In AI regulation, this could involve setting up mechanisms that allow regulators, innovators, and the public to contribute to and benefit from a shared pool of knowledge about AI’s impacts. This could include public databases of AI incidents, standardized reporting of AI-system performance, or platforms for sharing best practices in AI safety and ethics.

Static regulatory approaches may fail to capture the evolving landscape of AI applications and their societal implications. Instead, a dynamic, information-centric regulatory strategy that embraces the market as a discovery process could better facilitate beneficial innovations, while identifying and mitigating harms.

V. Conclusion

As the NTIA navigates the complex landscape of AI regulation, it is imperative to adopt a nuanced, forward-looking approach that balances the need to foster innovation with the imperatives of ensuring public safety and ethical integrity. The rapid evolution of AI technologies necessitates a regulatory framework that is both adaptive and principles-based, eschewing static snapshots of the current state of the art in favor of flexible mechanisms that could accommodate the dynamic nature of this field.

Central to this approach is to recognize that the field of AI encompasses a diverse array of technologies, methodologies, and applications, each with its distinct characteristics, capabilities, and implications for society. A one-size-fits-all regulatory model would not only be ill-suited to the task at-hand, but would also risk stifling innovation and hindering the United States’ ability to maintain its leadership in the global AI industry. NTIA should focus instead on developing tailored approaches that distinguish between low-risk and high-risk applications, ensuring that regulatory interventions are commensurate with the potential identifiable harms and benefits associated with specific AI use cases.

Moreover, the NTIA must resist the temptation to rely on overly prescriptive definitions of “openness” or to favor particular business models over others. The coexistence of open and closed approaches to AI development is essential to foster a vibrant, competitive ecosystem that drives technological progress and maximizes social benefits. By embracing a flexible regulatory framework that allows for experimentation and iteration, the NTIA can create an environment conducive to innovation while still ensuring that appropriate safeguards are in place to mitigate potential risks.

Ultimately, the success of the U.S. AI industry will depend on the ability of regulators, developers, researchers, and the broader community to collaborate in developing governance frameworks that are both effective and adaptable. By recognizing the importance of open development and diverse business models, the NTIA can play a crucial role in shaping the future of AI in ways that promote innovation, protect public interests, and solidify the United States’ position as a global leader in this transformative field.

[1] Dual Use Foundation Artificial Intelligence Models With Widely Available Model Weights, Docket No. 240216-0052, 89 FR 14059, National Telecommunications and Information Administration (Mar. 27, 2024) at 14063, question 8(a) [hereinafter “RFC”].

[2] See, e.g., Kristian Stout, Systemic Risk and Copyright in the EU AI Act, Truth on the Market (Mar. 19, 2024), https://truthonthemarket.com/2024/03/19/systemic-risk-and-copyright-in-the-eu-ai-act.

[3] Exec. Order No. 14110, 88 F.R. 75191 (2023), https://www.federalregister.gov/documents/2023/11/01/2023-24283/safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence?_fsi=C0CdBzzA [hereinafter “EO”].

[4] See, e.g., EO at §§ 1; 2(c), 5.2(e)(ii); and § 8(c);

[5] Id. at § 3(k).

[6] Id. at § (k)(iii).

[7] Id. at § 4.6. As NTIA notes, the administration refers to “widely available model weight,” which is equivalent to “open foundation models” in this proceeding. RFC at 14060.

[8] For more on the “open” vs “closed” distinction and its poor fit as a regulatory lens, see, infra, at nn. 19-41 and accompanying text.

[9] Adaptive regulatory frameworks are discussed, infra, at nn. 42-53 and accompanying text.

[10] See Steven C. Salop & David T. Scheffman, Raising Rivals’ Costs, 73:2 Am. Econ. R. 267, 267–71 (1983), http://www.jstor.org/stable/1816853.

[11] See Steven C. Salop & David T. Scheffman, Cost-Raising Strategies, 36:1 J. Indus. Econ. 19 (1987), https://doi.org/10.2307/2098594.

[12] Cindy Gordon, Google Pauses Gemini AI Model After Latest Debacle, Forbes (Feb. 29, 2024), https://www.forbes.com/sites/cindygordon/2024/02/29/google-latest-debacle-has-paused-gemini-ai-model/?sh=3114d093536c.

[13] Id.

[14] Id.

[15] Breck Dumas, Google Loses $96B in Value on Gemini Fallout as CEO Does Damage Control, Yahoo Finance (Feb. 28, 2024), https://finance.yahoo.com/news/google-loses-96b-value-gemini-233110640.html.

[16] RFC at 14060.

[17] RFC at 14062, question 1.

[18] RFC at 14062, question 3(a).

[19] Department of Commerce, Competition in the Mobile Application Ecosystem (2023), https://www.ntia.gov/report/2023/competition-mobile-app-ecosystem (“While retaining appropriate latitude for legitimate privacy, security, and safety measures, Congress should enact laws and relevant agencies should consider measures (such as rulemaking) designed to open up distribution of lawful apps, by prohibiting… barriers to the direct downloading of applications.”).

[20] RFC at 14061 (“‘openness’ or ‘wide availability’ of model weights are also terms without clear definition or consensus. There are gradients of ‘openness,’ ranging from fully ‘closed’ to fully ‘open’”).

[21] See Jonathan M. Barnett, The Host’s Dilemma: Strategic Forfeiture in Platform Markets for Informational Goods, 124 Harv. L. Rev. 1861, 1927 (2011).

[22] Id. at 2.

[23] Id. at 3.

[24]  Desktop Operating System Market Share Worldwide Feb 2023 – Feb 2024, statcounter, https://gs.statcounter.com/os-market-share/desktop/worldwide (last visited Mar. 27, 2024).

[25]  Andrei Hagiu, Proprietary vs. Open Two-Sided Platforms and Social Efficiency (Harv. Bus. Sch. Strategy Unit, Working Paper No. 09-113, 2006).

[26] Joey Sneddon, More Developers Use Linux than Mac, Report Shows, Omg Linux (Dec. 28, 2022), https://www.omglinux.com/devs-prefer-linux-to-mac-stackoverflow-survey.

[27] See Michael L. Katz & Carl Shapiro, Systems Competition and Network Effects, 8 J. Econ. Persp. 93, 110 (1994), (“[T]he primary cost of standardization is loss of variety: consumers have fewer differentiated products to pick from, especially if standardization prevents the development of promising but unique and incompatible new systems”).

[28] See. e.g., Nokia, Threat Intelligence Report 2020 (2020), https://www.nokia.com/networks/portfolio/cyber-security/threat-intelligence-report-2020; Randal C. Picker, Security Competition and App Stores, Network Law Review (Aug. 23, 2021), https://www.networklawreview.org/picker-app-stores.

[29] Collective Constitutional AI: Aligning a Language Model with Public Input, Anthropic (Oct. 17, 2023), https://www.anthropic.com/news/collective-constitutional-ai-aligning-a-language-model-with-public-input.

[30] RFC at 14061.

[31] Encryption and the “Going Dark” Debate, Congressional Research Service (2017), https://crsreports.congress.gov/product/pdf/R/R44481.

[32] EO at. § 3(k)(iii).

[33] EO at § 3(b).

[34] EO at § 4.2 (requiring companies developing dual-use foundation models to provide ongoing reports to the federal government on their activities, security measures, model weights, and red-team testing results).

[35] An End-to-End Platform for Machine Learning, TensorFlow, https://www.tensorflow.org (last visited Mar. 27, 2024).

[36] Learn the Basics, PyTorch, https://pytorch.org/tutorials/beginner/basics/intro.html (last visited Mar. 27, 2024).

[37] Akmalbek Bobomirzaevich Abdusalomov, Mukhriddin Mukhiddinov, & Taeg Keun Whangbo, Brain Tumor Detection Based on Deep Learning Approaches and Magnetic Resonance Imaging, 15(16) Cancers (Basel) 4172 (2023), available at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10453020.

[38] Keith T. Butler, et al., Machine Learning for Molecular and Materials Science, 559 Nature 547 (2018), available at https://www.nature.com/articles/s41586-018-0337-2.

[39] The Peltzman Effect, The Decision Lab, https://thedecisionlab.com/reference-guide/psychology/the-peltzman-effect (last visited Mar. 27, 2024).

[40] European Parliament, European Parliament legislative Resolution of 13 March 2024 on the Proposal for a Regulation of the European Parliament and of the Council on Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts, COM/2021/206, available at https://www.europarl.europa.eu/doceo/document/TA-9-2024-0138_EN.html [hereinafter “EU AI Act”].

[41] Id. at Art. 3(65).

[42] See Stephen G. Gilles, On Determining Negligence: Hand Formula Balancing, the Reasonable Person Standard, and the Jury, 54 Vanderbilt L. Rev. 813, 842-49 (2001).

[43] See Open Loop’s First Policy Prototyping Program in the United States, Meta, https://www.usprogram.openloop.org (last visited Mar. 27. 2024).

[44] Id.

[45] Justin (Gus) Hurwitz & Geoffrey A. Manne, Pigou’s Plumber: Regulation as a Discovery Process, SSRN (2024), available at https://laweconcenter.org/resources/pigous-plumber.

[46] Id. at 32.

[47] Id. at 33.

[48] See id. at 28-29

[49] Id. at 37.

[50] Id. at 37-38.

[51] Id.

Given how important digital connections are to Americans’ daily lives, it’s urgent that Congress move to renew the Federal Communications Commission’s authority to auction parts of the public airwaves.

That authority lapsed a little over a year ago and efforts to reinstate it have been repeatedly stuck in partisan gridlock.

Read the full piece here.

The European Parliament’s approval last week of the AI Act marked a significant milestone in the regulation of artificial intelligence. While the law’s final text is less alarming than what was initially proposed, it nonetheless still includes some ambiguities that could be exploited by regulators in ways that would hinder innovation in the EU. 

Among the key features emerging from the legislation are its introduction of “general purpose AI” (GPAI) as a regulatory category and the ways that these GPAI might interact with copyright rules. Moving forward in what is rapidly becoming a global market for generative-AI services, it also bears reflecting on how the AI Act’s copyright provisions contrast with current U.S. copyright law. 

Read the full piece here.

2024 will be a make-or-break year for the $42 billion taxpayer-funded Broadband Equity, Access, and Deployment program administered by the National Telecommunications and Information Administration.

Making BEAD a success and achieving the Biden Administration’s vision of “Internet for All” requires an ambitious “all-of-the-above approach” from federal, state, and local policymakers to take much-needed action on utility pole reforms – an often-neglected issue that is critical to ensuring unserved, rural communities are connected to reliable, high-speed internet.

Read the full piece here.

The Federal Trade Commission (FTC) has now gone on record in comments to the National Institute of Standards and Technology (NIST) that it supports expanded “march-in rights” under the Bayh-Dole Act (Act). But if NIST takes the FTC’s (unexpected, but ultimately unsurprising) contribution seriously, such an expansion could lead to overregulation that would ultimately hurt consumers and destroy the incentives that firms have to develop and commercialize lifesaving medicines.

Read the full piece here.

I. Introduction

This comment is submitted in response to the National Institute of Standards and Technology’s (NIST) request for information (RFI) on the Draft Interagency Guidance Framework for Considering the Exercise of March-In Rights.[1]

The U.S. patent system has been a major driver of innovation, and provides an important foundation for the nation’s technological leadership around the world. Undoubtedly, there are cases at the margins where one could find some invention has not been optimally commercialized. But the measure of the system’s success is not in isolated anecdotes, but rather, in data demonstrating it has been a major driver of economic growth and consumer welfare—both in general and particularly in the consistent development of lifesaving and life-enhancing medicines and medical devices.

This suggests that the integrity of the current patent rights framework under the Bayh-Dole Act is crucial for sustaining innovation, promoting commercialization, and ultimately enhancing consumer welfare. As such, any proposal to expand “march-in rights” must be treated with caution.

Further, while the administration’s focus in this draft guidance appears to be centered primarily on the pharmaceutical sector,[2] the proposed modifications have the potential to trigger extensive spillover effects across various other patent-reliant industries. For instance, industries such as biotechnology, software development, and advanced manufacturing—which rely fundamentally on strong patent protections to secure investments for research and development—could face unforeseen challenges. These sectors are driven by innovation underpinned by intellectual property. Increased uncertainty regarding the longevity and security of patent rights could lead them to experience a slowdown in the pace of that innovation, as venture capitalists may become more reluctant to fund new ventures. Of particular concern is that march-in petitions brought under a more liberal standard may become a useful tool for firms looking to stymie their competition.

The proposed changes are clearly unnecessary, given the history of success that characterizes the post-Bayh-Dole era. Indeed, these suggested modifications threaten to undermine a substantial portion of the U.S. economy and to harm both consumer health and general welfare. Apart from being ill-advised from an economic perspective, the proposed changes also appear to be at odds with the Bayh-Dole Act’s very legal and policy basis. As Adam Mossoff has observed, “the text of the Bayh-Dole Act and its consistent interpretation by federal officials militates against” the view that it authorizes imposing price controls on patented inventions produced with support from federal funding.[3]

In summary, the ongoing debate about modifying march-in rights under the Bayh-Dole Act touches on fundamental aspects of innovation, economic growth, and public welfare. This is not merely about adjusting a legislative framework; it is about preserving the delicate balance that has propelled the United States to the forefront of global innovation, particularly in life-saving pharmaceuticals and technologies. Any alterations to the Act’s implementation risk distorting this balance, potentially stifling innovation and undermining the economic and health benefits that have been realized. As such, it is imperative to carefully consider any proposed modifications to ensure that they support, rather than hinder, the Act’s foundational goal of fostering innovation and delivering tangible benefits to society.

II. Success of the Bayh-Dole Act and the Importance of Patent Rights

The Bayh-Dole Act, formally known as the University and Small Business Patent Procedures Act of 1980 (Act),[4] is a landmark piece of intellectual-property legislation. The Act allows universities, small businesses, and nonprofit organizations to retain and exercise patent rights to inventions developed under federally funded research programs. This legislative framework was designed to:

• Facilitate the transfer of federally funded research from academic and research institutions to the private sector for further development and commercialization;
• Encourage the practical application of these inventions for public benefit;
• Stimulate collaboration between public research entities and the private sector; and
• Enhance the contribution of federally funded inventions to the market, thereby boosting economic growth and public welfare.[5]

The Act has been a pivotal catalyst in advancing U.S. technological innovation, primarily by establishing a property-rights framework that creates incentives for the commercialization of scientific developments that received some degree of government funding. These property rights empower entities to license their inventions for more extensive applied research and development, thereby enhancing their accessibility and application for the broader public good.

The Act has been paying dividends since its inception in 1980. One important effect has been that, by enabling private companies to benefit from R&D that they (co-)fund at publicly supported universities, it has led to a dramatic increase in private-sector sponsorship of R&D at such universities. A report from the General Accounting Office (now known as the U.S. Government Accountability Office) found that, between 1980 and 1985 alone:

total business sponsorship of university research grew 74 percent, from $277 million in fiscal year 1980 to $482 million in fiscal year 1985 (in constant 1982 dollars). For 23 of the 25 universities we surveyed… industrial sponsorship of research more than doubled from $70 million in fiscal year 1980 to $160 million in fiscal year 1985 (in constant 1982 dollars).[6]

The Association of University Technology Managers (AUTM) estimates that, between 1996 and 2010, academic licensors contributed between $86 billion and $338 billion to U.S. gross domestic product (in 2005 dollars), in addition to supporting between 900,000 and 3 million person-years of employment over that that period.[7] In a survey of the 2019-2020 period, AUTM found that innovations of the sort that are at the core of the Bayh-Dole Act’s focus led to a 7% increase in startups; a 7% increase in invention disclosures; an 11% increase in net patent applications; a 3% increase in licenses executed; and a 31% increase in new products introduced to market based on academic research.[8]

Along with many other pro-innovation policies enacted over the last several decades, one of the Act’s enduring legacies is the fundamental shift it initiated in relocating innovative activity from Europe and Asia to the United States, with the latter now firmly established as the most important locale for producing new medicines:

In the last decade, while the U.S. had 111 [new chemical entities] discovered, Switzerland-headquartered companies were second with 26. This means that actual [new chemical entities] discovered that had a significant U.S. nexus for research and development is much higher than the 57 percent of total [new chemical entities] discovered, perhaps closer to 65 percent. One other point worth noting… is the reduction in overall [new chemical entities] discovered from the decade of the 1980s to now. The U.S. has the vast majority of clinical trials. A similar trend has taken place for medical devices.[9]

The United States has continued to develop a large number of new chemical entities in absolute terms, and in relative terms, has come to completely dominate the field.[10] This boom of patented innovations has also given rise to numerous transformative products we now consider commonplace, such as various cancer treatments,[11] prosthetics and medical devices,[12] a variety of web technologies, and improved foods.[13]

Nevertheless, the Act and the patent system are not without critics. Some have challenged the idea that the patent system does not sufficiently stimulate the production of inventions at universities,[14] or that, when such inventions occur, “large portion of those royalties… are derived from a few sizeable inventions at a handful of academic institutions.”[15] Thus, according to these critics, the Act does not promote widespread welfare gains, so much as enable large gains to a small number of parties.

Proposed changes to federal policy have also threatened to pare back the gains the Act has helped to facilitate. In addition to this draft guidance, which would introduce de facto price controls on any industry substantially reliant on patented invention, the U.S. Energy Department has been imposing more stringent domestic-manufacturing requirements on licensees—an obligation that makes little sense in our globalized economy and that is more likely to impose red tape without substantially improving domestic production.[16] In a 2021 letter to the Pentagon, Sen. Elizabeth Warren (D-Mass.) and Rep. Lloyd Doggett (D-Texas) noted that “[r]ecognizing the high prices of medical products developed, in part, with DOD funding, the Senate Armed Services Committee directed DOD to utilize march-in rights to lower prices.”[17] That is to say, at least some members of Congress have called explicitly for diminution of property rights and imposition of price controls.

But critics of the current patent system take far too dim a view of the Bayh-Dole Act’s legacy. Both the patent system and the Act provide important incentives not just to spur invention, but also to encourage commercialization. As noted above, the Act has performed remarkably well at opening opportunities for the commercialization of inventions, and it is this commercialization function that helps to ensure that crucial discoveries are not left to gather dust. Indeed, one of the main drivers of the Act’s success is its harmony with the economic theory of patent rights.

A. The Centrality of Strong Patent Protections

The biotechnology sector historically has depended on patents as a means to organize collaboration among universities, startups, and larger corporations. The costly and complex process of moving a discovery from the laboratory to the marketplace depends heavily on the temporary exclusivity granted by patent rights, as well as the data-protection rights of biologics subject to regulatory approval.[18] Such property rights are fundamental for attracting investors to commit resources to these ventures, which are fraught with high risks and significant costs.

Nobel laureate Kenneth Arrow observed that the product of inventive activity is knowledge.[19] This distinguishes knowledge from other goods or services, in that knowledge is costly to produce, but nearly costless to distribute.[20] In addition, information is often indivisible.[21] Indivisibility means that the information cannot be divided or allocated across producers, products, or outputs—e.g., once a drug’s chemical structure is known, this knowledge does not vary with how many doses are produced, or who produces them.[22] In addition, unlike most products and services, once knowledge is obtained, it is known forever. Those in possession of it can often utilize it with relatively little or no further expenditure. While a bicyclist may need to buy a new bicycle, his knowledge of how to ride will, once acquired, remain with him throughout his life. Likewise, once one knows how to produce a new drug, copies can often be reproduced at relatively low cost.

Another feature of knowledge is that consumers may not know its value until considerable resources have been expended to uncover it.[23] Consumers of a new drug do not know its safety and efficacy of until investigations have established how it performs biologically, which requires extensive modeling, as well as animal and human trials—the latter of which is especially costly.[24]

All these factors place drugs in the category of goods that are expensive to research, develop, and bring to market, but relatively cheap to imitate, as explained by Kip Viscusi and his co-authors:

Suppose the inventor discovers an important drug, Panacea. The inventor could keep the chemical structure secret and try selling the drug as a cure for certain diseases. But a rival could easily buy a few pills, hire a chemist to figure out the structure, and begin selling exact copies at a lower price.[25]

These rivals would benefit from the inventor’s investment in researching the new discovery at little expense of their own. In what is likely the most-cited empirical research on imitation costs, Edwin Mansfield et al. find that 6o% of the patented innovations in their sample were imitated much more quickly and at much lower cost than the initial innovation:

In the ethical drug industry [i.e., the part of the industry involved in researching, developing and bringing drugs to market with regulatory approvals], patents had a bigger impact on imitation costs than in the other industries, which helps to account for survey results indicating that patents are regarded as more important in ethical drugs than elsewhere. … Without patent protection, it frequently would have been relatively cheap (and quick) for an imitator to determine the composition of a new drug and to begin producing it. However, for many of these electronics and machinery innovations, it would have been quite difficult for imitators to determine from the new product how it is produced, and patents would not add a great deal to imitation cost (or time).[26]

If the benefits of the costly investment can be easily appropriated by rivals, then the incentives for invention evaporate. This leads to reduced investment, as explained in a section titled “Imitation Discourages Research” in Dennis Carlton and Jeffrey Perloff’s textbook:

Without a patent, anyone could use new information and imitations of new inventions could be sold legally. Suppose you discovered a cure for AIDS. You could sell your new drug for large sums of money if a patent gave you exclusive rights. Without a patent, other companies could duplicate your drug, and competition would drive the price to the competitive level. You would incur all the research costs, but not all the private benefit.[27]

Commenting on a 1990s-era proposal to regulate the pricing of “breakthrough” drugs, Viscusi et al. conclude that the proposal would ripple through companies’ R&D portfolios:

If one regards R&D investment as somewhat like a lottery—with low probabilities of achieving huge returns—top decile regulation changes completely the nature of the game. Winning the lottery now provides only a reasonable or breakeven return, with other outcomes worse![28]

Not only would such regulations affect companies’ expected returns, but they would also increase the variation in those returns. The added regulatory uncertainty would reduce firms’ confidence in the reliability of their return-on-investment projections. Because of the well-known and widely accepted risk-return tradeoff, firms that face increased uncertainty in investment returns will demand higher expected returns from the investments they pursue.[29] In other words, policies such as the proposed “march-in” rights simultaneously reduce expected investment returns and increase the required rate of return to invest in R&D, thereby reducing investment.

The history of patent commercialization supports the economic theory above. Prior to enactment of the Bayh-Dole Act, the federal government had a patchwork of often-stringent requirements on patenting and licensing agreements for projects it had funded.[30] The result was that many firms were hesitant to make large investments in the basic discoveries that were necessary to create commercial products.[31] Indeed, this makes sense, as a key feature of the patent system is that it can ensure the stability needed to attract investment and the large-scale diffusion of innovations across the market.

The evidence abundantly demonstrates that robust property-rights systems have been crucial to economic growth and prosperity.[32] These rights facilitate specialization and trade, which lead to innovation and growth. Intellectual property plays a crucial role in this dynamic. While there may be debates over the exact parameters of any patent-protection regime, strong evidence supports the idea that robust patent protection is vital for economic growth. Stephen Haber highlights that enforceable patent rights correlate with significant GDP increases.[33] Patricia Schneider’s research indicates that intellectual property substantially fosters innovation in developed countries.[34] Similarly, Yee Kyoung Kim and colleagues conclude that intellectual property boosts innovation.[35] Theoretical work by Daron Acemoglu and Ufuk Akcigit underscores the importance of patents, especially where inventors are significantly advanced technologically.[36] Yum Kwan and Edwin Lai suggest that inadequate intellectual-property protection causes greater welfare losses than does overprotection.[37]

Relatedly, Nobel laureate economist William Nordhaus has found that, even with patented discoveries, only a tiny fraction of the social returns from technological advancements is captured by producers, while the majority of benefits accrue to consumers.[38]

Patents are particularly important for startups, whose ability to exercise enforceable patent rights is key to market entry. There are three primary reasons for this: 1) injunctions protect startups from being copied by established firms, who might otherwise copy startups’ discoveries and pay court-set royalties; 2) patents serve as collateral to secure startup funding; and 3) patents attract venture-capital investment.

Diminishing patent rights by removing exclusion rights would allow larger firms to imitate startup innovations, reinforcing their market dominance. Without the threat of copying, established companies are forced to either innovate independently or acquire innovative startups. This aspect is particularly crucial for startups, as it protects their inventions from being misappropriated by larger rivals. The literature on firms’ strategies to prevent rivals from copying their inventions suggests that, while patents are not the only method, they are crucial in certain industries, most notably in pharmaceuticals and chemicals. [39]

Another key aspect of strong intellectual property rights is that they can allow firms to raise funds through the process of collateralization. This is particularly relevant for startups that lack tangible assets, as they can offer patents as security for funding.[40] As Gaétan de Rassenfosse puts it:

SMEs can leverage their IP to facilitate R&D financing…. [P]atents materialize the value of knowledge stock: they codify the knowledge and make it tradable, such that they can be used as collaterals. Recent theoretical evidence by Amable et al. (2010) suggests that a systematic use of patents as collateral would allow a high growth rate of innovations despite financial constraints.[41]

But the complexity in valuing patents,[42] particularly in the face of infringement risks, underscores why reliable IP rights are so important to maintaining patents’ value as collateral. As Jayan Kumar observes (in the parallel context of copyright):

Infringement action (most obviously music piracy) can seriously erode revenue streams and plans for combating infringement through litigation must be in place in order to protect the value of IP. Given the above risks and complexities, due diligence on IP before securitization is more expensive than with traditionally securitized assets.[43]

This last point becomes crucial to consider for the draft guidance, given that liberalizing march-in rights will almost certainly lead to increased litigation exposure across all industries that rely on patented technologies.

Lastly, as suggested above, intellectual-property protection influences venture-capital activity significantly. Patents impede imitation, can be used as collateral, and can help facilitate specialization, thereby fostering the entry of new specialized firms. Additionally, patents often signal to investors a company’s potential success and value. Empirical studies show that patent filings have significant positive effects on investor valuations, especially for early-stage companies, and play an important role as a “commitment device,” protecting entrepreneurs from investor expropriation.  For example, David Hsu and Rosemarie Ziedonis find:

a statistically significant and economically large effect of patent filings on investor estimates of start-up value…. A doubling in the patent application stock of a new venture [in] this sector is associated with a 28 percent increase in valuation, representing an upward funding-round adjustment of approximately $16.8 million for the average start-up in our sample.[44]

They also note that the effect is more pronounced in earlier financing rounds, when uncertainty surrounding the value of the underlying company is greater.[45]  Along similar lines, Carolin Häussler, Dietmar Harhoff, and Elisabeth Mueller show that “companies’ patenting activities have consistent and cogent effects on the timing of VC financing. Having at least one patent application reduces the time to the first VC investment by 76%.”[46] Other authors argue that patents may serve as a commitment device to protect entrepreneurs from the risk of expropriation by their early investors.[47]

The conclusion is clear: intellectual property is a significant contributor to innovation and should be a central element of growth strategies. This view is widely accepted among economists, particularly in industries with very large upfront costs and steeply declining marginal costs of production—of which, pharmaceuticals is perhaps the most extreme example.

Having said that, it would be naïve to think that U.S. intellectual-property law has reached a state of perfection. Intellectual-property protection must strike a delicate balance between guarding knowledge that could otherwise be replicated at minimal cost—thereby encouraging the creation of such knowledge—and ensuring that the knowledge is disseminated to the public. Even a minor shift in that balance toward dissemination and away from protection could have disproportionate effects, making copying (i.e., free-riding on the innovations of others) a more attractive strategy. This could lead to underinvestment and economic stagnation. Thus, when thinking about making changes to the status quo, policymakers should proceed with utmost care. The world preeminence that the current U.S. patent system has helped bring to fruition could easily be destroyed.

III.    March-In Rights and the Danger to Innovation

The proposed changes to the Bayh-Dole Act’s march-in rights[48] pose serious threats to the successful innovation regime that has propelled the United States to the forefront of global innovation.  In particular, the proposed revisions would expand the criteria for federal agencies to exercise march-in rights, potentially allowing for broader interpretation and application. Most concerning is that the proposed framework would allow agencies to consider such factors as the pricing of commercial goods and services arising from federally funded inventions.[49] Tellingly, the proposed framework would grant agency regulators authority to determine when a price is “extreme and unjustified given the totality of circumstances” and to decide, on that basis, whether to exercise march-in rights.[50]

These proposed changes raise concerns about their potential impact on the incentives for private-sector investment in the commercialization of federally funded research. Such changes threaten to disrupt the delicate balance of incentives that the Bayh-Dole Act has successfully established for more than four decades, potentially hindering innovation and diminishing consumer welfare in the long run.

But more importantly, one fundamental flaw in the draft framework would return us to a pre-1980 status quo ante. One of the primary questions that needs to be brought into focus in this proceeding is: what method of price discovery leads to the optimal commercialization of new patented inventions? Since much of this proceeding is focused on pharmaceutical products, we will restrict our discussion to the pricing of these products. Much of the economics of pricing patented medicines, however, transfers well to other contexts involving patent protections. As we discuss below, regulators are fundamentally incapable of matching, on average, the market’s efficiency in setting prices.

To understand the pricing of new pharmaceuticals, it’s helpful to begin with standard neoclassical price theory. The most basic model assumes that patented pharmaceuticals establish a monopoly, and that the monopolist sets different prices for different consumers based on their willingness to pay. In principle, such a “price-discriminating monopolist” will charge each consumer a different price and the lowest price paid will be equal to the drug’s marginal cost of production. In other words, those consumers least willing to pay will pay the same price as in a “perfectly competitive” market. Moreover, the amount of the drug produced will be the same as under perfect competition. The big difference is that the producer receives all the consumer surplus. In practice, pharmaceutical companies are not perfectly discriminating monopolists, but they do typically set different prices in different countries and for different patient groups.[51]

In reality, very few—if any—new pharmaceuticals actually enjoy a monopoly. At best, they represent a new class of drug for treating a condition. Even in such cases, they typically compete with older products that are either less effective or have more side effects for some proportion of patients.[52] This competition introduces a dynamic interplay between the new and old products, influencing the innovator’s pricing strategy.

The neoclassical model shows that even a profit-maximizing monopolist has incentives to offer products at a range of prices to different consumers. But when the “monopolist” assumption is relaxed—reflecting the reality of competitive dynamics both within and between classes of drugs for any particular condition­–it becomes even more difficult, if not impossible, to determine whether a particular drug price is “extreme and unjustified.” There is thus a high likelihood that any such intervention would be arbitrary and capricious.

Unfortunately, if given such a mandate, regulators are likely to have incentives to intervene for political reasons. In essence, regulators gain little by declining to intervene in the presence of an alleged “extreme and unjustified” drug price.[53] Meanwhile, the consequences of (practically ubiquitous) improper intervention would not be borne by the regulator, but by the innovators and patients.

When a private firm misjudges demand and sets its prices incorrectly, it faces punishment by the market. This, in turn, leads the firm to correct its pricing strategy. Liberalized march-in rights, by contrast, create incentives for a one-way ratchet, whereby regulators—themselves insulated from market discipline—are driven by political pressures to demand price reductions, regardless of the effect on firms’ incentives to develop new medicines.

A.      Intrinsic Complexities

The economics of drug development and pricing in the pharmaceutical industry present unique challenges that set it apart from many other sectors. While the fundamental principles of the price system apply to patented inventions in this field, the intricacies of pharmaceutical development necessitate more complex pricing strategies.

One of the defining characteristics of pharmaceutical R&D is the very long time it takes to bring a drug to market. From initial discovery to market launch, the process of developing a new drug typically takes between 12 and 15 years.[54] This extended timeframe is due largely to the rigorous clinical trials and associated regulatory approvals that each new drug must undergo to ensure safety and efficacy. This prolonged development period represents a significant commitment of time and resources, often with no guarantee of success.[55]

Many potential drugs that enter the development pipeline do not make it to market, either due to inefficacy, safety concerns, or other factors discovered during the development process.[56] This high attrition rate means that successful drugs must not only cover their own development costs but also compensate for the expenses incurred by those that failed.[57] A 2016 study found that the likelihood of a molecule selected for clinical trials successfully concluding all three phases of trials and going to market is around 12%.[58] Taking into account this low success rate, the authors estimate the average cost of developing a new approved drug to be $2.8 billion.[59]

Given these unique challenges­—long development times, substantial upfront investments, and a high rate of failure—pharmaceutical pricing must be carefully calibrated. Pricing strategies must account for recouping large investments while also considering the competitive market landscape, regulatory environment, and patient access.

B.      Regulatory Complexities

The challenge is magnified when one considers the complex regulatory environment that exerts significant distortionary pressures on drug pricing. For example, there are several federal programs—including Medicaid,[60] the 340B Drug Pricing Program,[61] and the regulations for the coverage gap for Medicare Part D[62]—that impose price controls on pharmaceuticals. While these controls aim to make medications more affordable for certain groups, the challenges they inadvertently create for pharmaceutical companies include potential distortions of downstream pricing for drugs outside of these programs.

For example, among these policies’ unintended consequences is to penalize companies that offer drugs at lower prices. The mandated discounts and rebates for government programs often mean that pharmaceutical companies receive less revenue for the same product, relative to the open market.[63] To compensate for revenue losses incurred in these programs, pharmaceutical companies are often compelled to raise prices for patients not covered by these federal programs.[64] This situation creates a disparity in drug pricing, where the burden of subsidizing the cost for government programs falls indirectly on other consumers, often resulting in higher overall healthcare costs.

Furthermore, this regulatory thicket complicates drugmakers’ pricing strategies. Instead of pricing based strictly on market demand or research and development costs (which is complicated enough on its own), companies must navigate a maze of regulations and mandatory discounts. This distorts natural market dynamics, often leading to higher prices for some consumers to balance the reduced revenue from government-mandated pricing. This approach can also stifle innovation, as pharmaceutical companies may redirect resources from research and development to regulatory compliance and strategic-pricing management.

C.      The Fraught Nature of Intervening in Market-Based Drug Pricing

It’s worth noting that march-in rights have not, to date, been exercised. This fact serves as an implicit acknowledgment of the pharmaceutical industry’s effective functioning within the constraints noted above. Moreover, it reflects regulators’ prudent reluctance to intervene in a complex and delicately balanced ecosystem. Indeed, any intervention in such a nuanced sector runs the risk of arbitrariness, given the intricacies involved in drug development and pricing. The restraint regulators have shown underlines their understanding of the unique economic dynamics of the pharmaceutical industry and the potential unintended consequences of intervention.

Further, the economics of the pharmaceutical industry also reveal the role that successful, high-revenue drugs have played in cross-subsidizing those discoveries that generate lower revenues.[65] This interplay between different segments of a pharmaceutical company’s portfolio is another crucial factor that militates against pricing interventions. The inherent support that successful patented medicines offer to the research and development of less profitable drugs (and total failures) is a vital component of the industry’s ecosystem.

So-called “blockbuster” drugs are a boon not just for the pharmaceutical companies, but also for the broader healthcare system. Some of the profits from these successful drugs are reinvested into further research and development, fueling the discovery and production of new medications.[66] This cycle of profit and reinvestment is critical to sustain the development of drugs that may have a smaller absolute market but are vital for treating rarer conditions. In this way, the big winners in a pharmaceutical company’s portfolio underpin the development and continued availability of lower revenue drugs and experiments with seemingly promising, but ultimately unfruitful, lines of research.

Therefore, any intervention in pharmaceutical pricing must be approached with caution. The cross-subsidization model represents a delicate balance essential not just for pharmaceutical firms’ financial health, but also to ensure the availability of a wide range of medications that meet diverse health-care needs. Unfortunately, this balance has already been weakened by price controls both in the United States and internationally, and could be substantially harmed by new price controls or other regulatory interventions.

Intervening in the pharmaceutical industry’s complex, carefully balanced, intricate, and multifaceted domain of drug development and commercialization risks creating an environment in which outcomes are dictated by centralized agencies, rather than by decentralized, bottom-up processes. In such a system, regulators’ necessarily limited knowledge will inevitably result in inferior outcomes. Moreover, it will lead to picking winners and losers in an arbitrary and capricious manner.

The issue’s complexity is compounded by the fact that the vast majority of drugs that are developed receive some federal funding.[67] While it is impossible to know whether the same drugs would be developed without such funding, the fact is that such funding crowds out private investment in basic R&D. Moreover, it means that the proposed expansion of march-in rights would apply to nearly every patented drug currently on the market and in development. Therefore, such interventions would not only be arbitrary and capricious, in ways that raise constitutional questions, but also ominous and all-encompassing.

Moreover, the error costs associated with such interventions cannot be overlooked. In the pharmaceutical industry, the journey from lab to market is fraught with uncertainties and high failure rates. For instance, only a quarter of drugs that complete Phase 3 clinical trials proceed to Phase 4.[68] Reasons for this can include a lack of efficacy in larger populations or commercial non-viability.[69] A regulatory body attempting to override these decisions would need to possess better knowledge than the compound’s own developers and commercializers regarding what will ultimately prove viable in the market. This prospect is clearly absurd and would lead to misallocation of resources, with companies being perversely encouraged to chase a higher number of unsuccessful endeavors.

Thus, any regulatory intervention in this space must be undertaken with a deep understanding of the inherent complexities and uncertainties of drug development. A regulator’s decision to intervene in the commercialization process could result in significant wasted resources and could potentially impede the development of truly effective and needed medicines. The challenge lies in striking the right balance between encouraging innovation and ensuring access to effective and affordable medications, without falling into the trap of overregulation that could stifle progress in this vital field.

IV.    Conclusion

In short, the narrative that drives the conversation around altering march-in rights is deeply flawed. The Bayh-Dole Act does not unjustly deprive taxpayers of the innovations they partially funded through their contributions to the federal government. In fact, the Act has fostered an explosion of innovative activity that yields enormous benefits, both seen and unseen, to American consumers. The observable benefits are evident in the ever-expanding access to new medicines and devices that improve health outcomes for consumers.[70] The unseen—or rather, the easy to miss—benefits include the economic growth that has resulted from the United States serving as a major hub for innovative research and development.[71] The status quo is wildly successful and any perceived failures should be addressed with targeted solutions, not with a wholesale alteration to the framework that has been responsible for driving these changes.

Further, it’s crucial to understand the effects that expanding march-in rights to address instances of “extreme” pricing could have on the nature of the Act itself. Originally designed as a pro-innovation policy, the Bayh-Dole Act could inadvertently transform into a regulatory tool for market manipulation.

Regulations are often complex and challenging to navigate. This complexity creates opportunities for incumbent firms to leverage regulations to their advantage, and to the detriment of competition and consumer welfare.  In the context of the Bayh-Dole Act, expanding march-in rights to tackle “extreme” pricing could lead to just such a perverse outcome. Such a scenario would mark a significant shift from the Bayh-Dole Act original intent of fostering innovation toward a landscape where regulatory manipulation becomes a key competitive strategy. This potential transformation underscores the need for careful consideration and a balanced approach in any amendments to the Act. Addressing the issue of pricing should not compromise the Act’s ability to stimulate innovation and healthy market competition.

Finally, expanding march-in rights under the Bayh-Dole Act, although primarily targeted at pharmaceutical producers, sets a precedent with far-reaching implications for all patent-reliant industries, including computers, biotech, and manufacturing. Industries that thrive on intellectual property to develop and safeguard their innovations will be watching this development closely. This potential for regulatory and legal manipulation could alter the competitive landscape, where gaining an upper hand might no longer depend solely on innovation and market strategies, but increasingly on the ability to navigate and exploit expanded march-in rights.

[1] Request for Information Regarding the Draft Interagency Guidance Framework for Considering the Exercise of March-In Rights, 88 FR 85593 (Dec. 8, 2023), https://www.federalregister.gov/documents/2023/12/08/2023-26930/request-for-information-regarding-the-draft-interagency-guidance-framework-for-considering-the [hereinafter “RFI”]

[2] For example, five of the eight “scenarios” presented in the RFI focus on biotechnology.

[3]  Adam Mossoff, The False Promise of Breaking Patents to Lower Drug Prices, 97 St. John’s L. Rev. (forthcoming 2023) (manuscript at 18), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4348499.

[4] 35 U.S.C. § 200, et seq. (2011).

[5] Id. at § 202.

[6] U.S. General Accounting Office, Patent Policy Recent Changes in Federal Law Considered Beneficial, GAO Report No. RCED-87-44 (1987), available at https://www.gao.gov/products/rced-87-44.

[7] Lori Pressman et al., The Economic Contribution of University/Nonprofit Inventions in the United States: 1996–2010, Biotechnology Industry Organization (Jun. 20, 2012) at 13,  available at https://archive.bio.org/sites/default/files/Pressman%2520BIO%25202012%2520Final%2520r1%2520w%2520cover%2520sheet_0.pdf.

[8] Joseph Allen, A Pandemic Can’t Stop Bayh-Dole—But Politicians Might, IPWatchdog (Aug. 31, 2021), https://ipwatchdog.com/2021/08/31/pandemic-cant-stop-bayh-dole-politicians-might/id=137235.

[9] Shanker Singham, Improving U.S. Competitiveness; Eliminating Anti-Competitive Market Distortions, at 12 (Int’l Roundtable Trade & Competition Pol’y., Nov. 15, 2011), available at https://shankersingham.com/2019/10/05/on-improving-us-competitiveness.

[10] Id.

[11] See, e.g., Molecular Biomarkers Improve Treatment of Colorectal Cancers, AUTM (2008), https://autm.net/about-tech-transfer/better-world-project/bwp-stories/medical-diagnostic-predictors-of-therapy-response (last visited Feb. 1, 2024); 3-D Virtual Colonoscopies: Changing Attitudes, Reducing Cancer, AUTM, https://autm.net/about-tech-transfer/better-world-project/bwp-stories/3-d-virtual-colonoscopy (last visited Feb. 1, 2024).

[12] See, e.g., Increasing Mobility for Amputees, AUTM (2016), https://autm.net/about-tech-transfer/better-world-project/bwp-stories/all-terrain-knee-(1) (last visited Feb. 1, 2024); Innovative Bandage Saves Lives, AUTM (2008), https://autm.net/about-tech-transfer/better-world-project/bwp-stories/alphabandage (last visited Feb. 1, 2024); Cochlear Implant Brings Sound and Language to Thousands, AUTM (2006), https://autm.net/about-tech-transfer/better-world-project/bwp-stories/cochlear-implant (last visited Feb. 1, 2024).

[13] Honeycrisp: The Apple of Minnesota’s Eye, AUTM (2018), https://autm.net/about-tech-transfer/better-world-project/bwp-stories/honeycrisp-apple (last visited Feb. 1, 2024).

[14] See, e.g., Lisa Larrimore Oullette & Andrew Tutt, How Do Patent Incentives Affect University Researchers?, 61 Int’l Rev. L. & Econ. 1 (2020), https://doi.org/10.1016/j.irle.2019.105883.

[15] David Orozco, Assessing the Efficacy of the Bayh-Dole Act Through the Lens of University Technology Transfer Offices (ITOS), 21 N.C. J.L. & Tech. 115, 142 (2019)

[16] See Frequently Asked Questions (FAQs) for Applicants and Awardees of DOE Financial Assistance and R&D Contracts Regarding the Department’s Determination of Exceptional Circumstances (DEC) for DOE Science and Energy Technologies Issued in June of 2021, U.S. Department of Energy (2021), available at https://www.energy.gov/sites/default/files/2022-03/FAQs_03092022.pdf; see also Joseph Allen, DOE’s Misuse of Bayh-Dole’s ‘Exceptional Circumstances’ Provision: How Uniform Patent Policies Slip Away, IPWatchdog (May 26, 2022), https://ipwatchdog.com/2022/05/26/misuse-bayh-doles-exceptional-circumstances-provision-uniform-patent-policies-slip-away/id=149275.

[17] See Elizabeth Warren & Lloyd Doggett, Letter to the Secretary of Defense Regarding Reducing Drug Prices (Jul. 22, 2021), available at https://www.warren.senate.gov/imo/media/doc/Letter%20to%20DOD%20about%20Reducing%20Drug%20Prices%20Final%207.22.21.pdf.

[18] Dana P. Goldman, Darius N. Lakdawalla, & Tomas Philipson, The Benefits From Giving Makers Of Conventional ‘Small Molecule’ Drugs Longer Exclusivity Over Clinical Trial Data, 30 Health Affairs 1, 84-90 (2011), available at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3804334.

[19] Kenneth J. Arrow, Economic Welfare and the Allocation of Resources for Invention, at 609, in The Rate and Direction of Inventive Activity (R. R. Nelson, ed., 1962).

[20] See id. at 614 (“The cost of transmitting a given body of information is frequently very low.”).

[21] See id.. at 615.

[22] See id. (“[T]he use of information about production possibilities, for example, need not depend on the rate of production.”)

[23] Id.

[24] Joseph A. DiMasi, Henry G. Grabowski, & Ronald W. Hansen, Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs, 47 J. Health Econ. 20 (2016), https://pubmed.ncbi.nlm.nih.gov/26928437.

[25] W. Kip Viscusi, John M. Vernon & Joseph E. Harrington, Jr., Economics of Regulation and Antitrust (2d ed., 1995) at 832.

[26] Edwin Mansfield, Mark Schwartz, & Samuel Wagner, Imitation Costs and Patents: An Empirical Study, 91 Econ. J. 907, 913 (1981). [emphasis added]

[27] Dennis W. Carlton & Jeffrey M. Perloff, Modern Industrial Organization (4^th ed., 2005) at 532. For a numerical example, see, Richard A. Posner, Economic Analysis of Law (4^th ed., 1992) at 38.

[28] Viscusi, Vernon &  Harrington, Jr., supra n. 25, at 863.

[29] See Edwin J. Elton & Martin J. Gruber, Modern Portfolio Theory and Investment Analysis (4th ed, 1991).

[30]  See, e.g., Jonathan Barnett, The Great Patent Grab, in The Battle Over Patents: History and Politics of Innovation (Stephen H. Haber & Naomi R. Lamoreaux eds., Oxford University Press 2021), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3909528; Mossoff, supra n. 3, at 18-20 (“Government ownership of patents proved to stifle, rather than to promote distribution of new innovations.”)

[31] Id.

[32] Stephen Haber, Patents and the Wealth of Nations, 23 Geo. Mason L. Rev. 811, 811 (2016) (“There is abundant evidence from economics and history that the world’s wealthy countries grew rich because they had well-developed systems of private property”); see also, Zorina Khan & Kenneth L. Sokoloff, Institutions and Democratic Invention in 19th-Century America: Evidence from “Great Inventors” 1790-1930, 94 Am. Econ. Rev. 400 (2004); Josh Lerner, The Economics of Technology and Innovation: 150 Years of Patent Protection, 92 Am. Econ. Rev. 221 (2002); Albert G.Z. Hu & Ivan P.L. Png, Patent Rights and Economic Growth: Evidence from Cross-Country Panels of Manufacturing Industries, 65 Oxford Econ. Papers 675 (2013) (finding faster growth and higher value in patent-intensive industries in countries that improve the strength of patents); Bronwyn H. Hall & Rosemarie Ham Ziedonis, The Patent Paradox Revisited: An Empirical Study of Patenting in the US Semiconductor Industry, 1979-1995, 32 RAND J. Econ. 101, 125 (2001) (identifying “two ways in which the pro-patent shift in the U.S. legal environment appears to be causally related to the otherwise perplexing surge in U.S. patenting rates, at least in the semiconductor industry”); Nikos C. Varsakelis, The Impact of Patent Protection, Economy Openness and National Culture on R&D Investment: A Cross-country Empirical Investigation, 30 Res. Pol’y 1059, 1067 (2001) (“Patent protection is a strong determinant of the R&D intensity, and countries with a strong patent protection framework invest more in R&D.”); David M. Gould & William C. Gruben, The Role of Intellectual Property Rights in Economic Growth, in Dynamics Of Globalization & Development 209 (Satya Dev Gupta & Nanda K. Choudhry eds., 1997) (“The evidence suggests that intellectual property protection is a significant determinant of economic growth. These effects appear to be slightly stronger in relatively open economies and are robust to both the measure of openness used and to other alternative model specifications.”)

[33] Haber, supra note 32, at 816. (“Figure 1 therefore presents a graph of the strength of enforceable patent rights and levels of economic development for all non-petro states in 2010. There is nothing ambiguous about the resulting pattern: there are no wealthy countries with weak patent rights, and there are no poor countries with strong patent rights. Indeed… as patent rights increase, GDP per capita increases with it. Roughly speaking, for every one-unit increase in patent rights (measured from zero to fifty) per capita income increases by $780. A simple regression of patent rights and patent rights squared on GDP indicates that roughly three-quarters of the cross-sectional variance in per capita GDP around the world is explained by the strength of patent rights.”) (emphasis added); see also Ronald A. Cass & Keith N. Hylton, Laws Of Creation: Property Rights In The World Of Ideas 45-46 (2013) (discussing results of regression analysis providing evidence that “countries with stronger intellectual property rights tend to grow economically more than those with weak intellectual property rights.”)

[34] Patricia Higino Schneider, International Trade, Economic Growth and Intellectual Property Rights: A Panel Data Study of Developed and Developing Countries, 78 J. Dev. Econ. 529, 539 (2005) (“The results suggest that IPRs have a stronger impact on domestic innovation for developed countries. This variable is positive and statistically significant in all OLS regressions in Table 4 (developed countries).”)

[35] Yee Kyoung Kim, Keun Lee, Walter G. Park, & Kineung Choo, Appropriate Intellectual Property Protection and Economic Growth in Countries at Different Levels of Development, 41 Res. Pol’y 358, 367 (2012) (“[T]he impact of patenting intensity on growth is much larger in high income countries, as can be seen from the positive coefficient of the interaction term between the high income country dummy and patenting intensity – this coefficient being statistically significant at the 1% level of statistical significance. From column 6, the measured net effect of patent intensity on growth in high income countries is 0.0683 (=−0.027 + 0.953, where the former is the coefficient of the patenting intensity of middle-to-low-income countries and the latter the coefficient of the interaction term between the high income country dummy and patenting intensity).”)

[36] Daron Acemoglu & Ufuk Akcigit, Intellectual Property Rights Policy, Competition and Innovation, 10 J. Eur. Econ. Ass’n. 1, 1 (2012) (“[O]ptimal policy involves state-dependent IPR protection, providing greater protection to technology leaders that are further ahead than those that are close to their followers.”)

[37] Yum K. Kwan & Edwin L-C Lai, Intellectual Property Rights Protection and Endogenous Economic Growth, 27 J. Econ. Dynamics & Control 853, 854 (2003) (“The calibration results indicate that there is under-protection of IPR (relative to the optimal level) within plausible range of parameter values, and that under-protection of IPR is much more likely than over-protection. More complete computation indicates that in the case of over-protection, the welfare losses are trivial; whereas in the case of under-protection, the welfare losses can be substantial. One interpretation of this result is that the US should protect IPR much more than it currently does.”)

[38] William D. Nordhaus, Schumpeterian Profits in the American Economy: Theory and Measurement at 1 (Nat’l Bureau of Econ. Res. Working Paper No. 10433 Apr. 2004), http://www.nber.org/papers/w10433.

[39] See, e.g., Edwin Mansfield, Patents and Innovation: An Empirical Study, 32 Mgmt. Sci. 173, 175-176 (1986) (Mansfield shows through surveys that patent protection only had a limited impact on innovation in industries other than the pharmaceutical industry and, to a lesser extent, the chemical industry. Mansfield argues that this is because the effectiveness of patents depends on the extent to which they increase imitation costs; and that this increase is more substantial in the chemical and pharmaceutical industries). Note that this study largely predates standard-reliant industries, such as mobile-communications technology, where patents likely play a very important role in creating appropriability. See also Richard C. Levin, Alvin K. Klevorick, Richard R. Nelson, Sidney G. Winter, Richard Gilbert, & Zvi Griliches, Appropriating the Returns from Industrial Research and Development, 3 Brookings Papers On Econ. Activity 783, 797 (1987). Levin et al.’s findings are broadly in line with Mansfield’s. More recently, these findings were supported by Cohen et al. See Wesley M. Cohen, Richard R. Nelson, & John P. Walsh, Protecting Their Intellectual Assets: Appropriability Conditions and Why US Manufacturing Firms Patent (or Not) (Nat’l Bureau of Econ. Res. Working Paper 7552, Feb. 2000), https://www.nber.org/papers/w7552.

[40] See, e.g., Mario Calderini & Maria Cristina Odasso, Intellectual Property Portfolio Securitization: An Evidence Based Analysis, Innovation Studies Working Paper (ISWOP), NO. 1/08, at 33 (2008) (“[I]t seems that patent securitization should be more suitable for small and medium companies with a consistent IP portfolio but that have not easy access to capital market or have a higher financial risk and few possibility to raise unsecured financing.”); see also Dov Solomon & Miriam Bitton, Intellectual Property Securitization, 33 Cardozo Arts & Ent. L.J. 125, 171-73 (2015) (“Among the famous securitization transactions in the field of IP rights are the securitizations of the copyrights of the singer David Bowie, the trademark of the Domino’s Pizza chain, and the patent on the HIV drug developed by Yale University.”); Nishad Deshpande & Asha Nagendra, Patents as Collateral for Securitization, 35 Nature Biotechnology 514, 514 (2017) (“Patents are important assets for biotech organizations, not only for protecting inventions but also as assets to raise monies.”); Tahir M. Nisar, Intellectual Property Securitization and Growth Capital in Retail Franchising, 87 J. Retailing 393, 393 (2011) (“A method of raising finance particularly suited to retail franchisors is intellectual property (IP) securitization that allows companies to account for intangible assets such as intellectual property, royalty and brands and realize their full value. In recent years, a number of large restaurant franchisors have securitized their brands to raise funds, including Dunkin Brands and Domino’s Pizza (Domino’s). We use property rights approach to show that IP securitization provides mechanisms that explicitly define ownership of intangible assets within the securitization structure and thus enables a company to raise funds against these assets.”)

[41] Gaétan De Rassenfosse, How SMEs Exploit Their Intellectual Property Assets: Evidence from Survey Data, 39 Small Bus. Econ. 437, 439 (2012).

[42] See Solomon & Bitton, supra note 40 (discussing the difficulties in evaluating patents as a barrier to securitization); see also Aleksandar Nikolic, Securitization of Patents and Its Continued Viability in Light of the Current Economic Conditions, 19 Albany L.J. Sci. & Tech. 393, 491 (2009) (“Anyone attempting to accurately assess the value of a patent portfolio faces numerous challenges including potential invalidity proceedings, potential infringement and infringement proceedings, obsolescence, or lack of demand for a license or the invention itself.”)

[43] Jayant Kumar, Intellectual Property Securitization: How Far Possible and Effective, 11 J. Intellectual Prop. Rights 98, 98 (2006).

[44] David H. Hsu & Rosemarie H. Ziedonis, Patents as Quality Signals for Entrepreneurial Ventures, Acad. Mgmt. Proceedings, Vol. 1, at 6 (2008), available at https://faculty.wharton.upenn.edu/wp-content/uploads/2015/07/11.pdf.

[45] Id.

[46] Carolin Häussler, Dietmar Harhoff & Elisabeth Müller, To Be Financed or Not… — The Role of Patents for Venture Capital-Financing, at 3 (ZEW-Centre for European Economic Research Discussion Paper 09-003, Mar. 28, 2013), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1393725; see also De Rassenfosse, supra note 41, at 441.

[47] See Ronald J. Mann & Thomas W. Sager, Patents, Venture Capital, and Software Start-Ups, 36 Research Pol’y 193, 207 (2007). (“We note one additional possibility suggested by the data, that portfolio firms obtain the patents not because they increase the value of the firm to its investors, but because they protect the contributions of the firm from expropriation by the investors. The idea here is that by giving the portfolio firm a cognizable property right in its technology, the patents increase the value of the firm by decreasing the costs of moral hazard and hold-up in the relations between the entrepreneurs and their investors. Shane (2002) proposes a similar mechanism to explain patterns in licensing of patents assigned to MIT.”)

[48] 35 U.S.C. 203 allows for a limited number of conditions under which federal agencies can grant licenses to inventions at least partially funded by federal money. These conditions include when a contractor or assignee is not expected to commercialize an invention in a reasonable amount of time, or when health or safety concerns are not expected to be reasonably satisfied by a contractor or assignee. Id. at (a)(1)-(2). To date, march-in rights have never been exercised. It should also be noted that “price” is not mentioned anywhere in § 203 as a basis for “march in,” which could lead to the possibility of a valid Supreme Court challenge to such a change under the “major questions doctrine.” See, e.g., The Major Questions Doctrine, CRS Report No. IF12077 (Nov. 2, 2022), https://crsreports.congress.gov/product/pdf/IF/IF12077. (“Under the major questions doctrine, the Supreme Court has rejected agency claims of regulatory authority when (1) the underlying claim of authority concerns an issue of “vast ‘economic and political significance,’” and (2) Congress has not clearly empowered the agency with authority over the issue.”) Moreover, the claim that the act was intended to be used to impose price controls is, at best, a stretch of statutory interpretation and, more realistically, a completely ill-fated enterprise that depends on taking statutory terms out of context. See Mossoff, supra n. 3, at 22-33.

[49] RFI at 85599.

[50] Id.

[51] See Paul Krugman & Robin Wells, Economics (4^th ed., 2015) at 391 (Regarding pricing of patent-protected drugs, “A monopolist will maximize profits by charging a higher price in the country with a lower price elasticity (the rich country) and a lower price in the country with a higher price elasticity (the poor country). Interestingly, however, drug prices can differ substantially even among countries with comparable income levels.”)

[52] For example, the H2 antagonist Tagamet (cimetidine) was developed by Smith, Kline & French to prevent and treat gastroesophageal reflux disease (GERD) and gastric ulcers. In response, Glaxo developed a similar but more effective H2 antagonist, Zantac (ranitidine) (See Viscusi et al., supra note 25 at 851-852). This within-class competition was followed by the development of a new, more-effective, and longer-lasting class of anti-GERD drugs known as proton-pump inhibitors (PPI), starting with omeprazole and soon followed by a slew of others, including lansoprazole and pantoprazole. See Daniel S. Strand, Daejin Kim, & David A. Peura1, 25 Years of Proton Pump Inhibitors: A Comprehensive Review, 15 Gut Liver. 11(1), 27-37 (2017), available at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5221858. The development of treatments for Alzheimer’s has followed a similar trajectory. Biogen’s Aduhelm (aducanumab), was recently retired, but the drug, which works by clearing the amyloid plaques that block neurotransmission in people with Alzheimer’s, has been hailed as a “groundbreaking discovery that paved the way for a new class of drugs and reinvigorated investments in the field.” See Editorial Board, Requiem for an Alzheimer’s Drug, Wall St. J. (Jan. 31, 2024), https://www.wsj.com/articles/aduhelm-biogen-alzheimers-treatment-drug-development-pharma-fda-1d866bd7. The development of Aduhelm thus served as both a foundation for other drugs in the same class of anti-amyloid monoclonal antibody treatments, such as Leqembi (lecanemab) (see Christopher H. Van Dyck et al., Lecanemab in Early Alzheimer’s Disease, 388 N. Engl. J. Med. 9-21 (2023), https://www.nejm.org/doi/full/10.1056/NEJMoa2212948) as well as continued within-class competition for those later drugs, until its retirement. Similarly, Cognex (tacrine)—the first in an earlier class of ameliorative drugs for Alzheimer’s (acetylcholineesterase inhibitors, AChEIs), which work by preventing the breakdown of the neurotransmitter acetylcholine—was, like Aduhelm, ultimately deemed relatively ineffective and withdrawn (See Nawab Qizilbash et al., WITHDRAWN: Tacrine for Alzheimer’s Disease, 18 Cochrane Database Sys. Rev. 3 (2007), https://pubmed.ncbi.nlm.nih.gov/17636619) because it had been superseded by other AChEIs, such as Aricept (donepezil). See Sharon L. Rogers et al., Donepezil Improves Cognition and Global Function in Alzheimer Disease, 158(9) Arch Intern Med. 1021-1031 (1998), available at https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/205223.

[53] See, e.g., Eric Fruits, The Oregon Health Plan: A “Bold Experiment” That Failed (Cascade Policy Institute, Sep. 2010), https://ssrn.com/abstract=1680047 (describing how covered treatments under Oregon’s Medicaid program was originally based on objective “cost-effectiveness” criteria, but quickly transitioned to subjective criteria based on public pressure).

[54] AI’s Potential to Accelerate Drug Discovery Needs a Reality Check, Nature (Oct. 10, 2023), https://www.nature.com/articles/d41586-023-03172-6.

[55] Duxin Sun, Wei Gao, Hongxiang Hu, & Simon Zhou, Why 90% of Clinical Drug Development Fails and How to Improve It?, 12 Acta Pharm. Sin. B 3049 (Jul. 2022); see also, Krugman & Wells, supra note 51 at 264 (“there is a huge failure rate along the way, as only one in five drugs tested on humans ever makes it to market.”)

[56] Sun et al., supra note 55.

[57] Research and Development in the Pharmaceutical Industry, Congressional Budget Office (Apr. 2021), https://www.cbo.gov/publication/57126 (“For established drug companies, current revenue streams from existing products also provide an important source of financing for their R&D projects.”)

[58] DiMasi, Grabowski, & Hansen, supra n. 24.

[59] Id.; see also, CBO, supra note 57 (“average R&D expenditures per new drug range from less than $1 billion to more than $2 billion”).

[60] See The Medicaid Prescription Drug Rebate Program, established by the Omnibus Budget Reconciliation Act (OBRA) of 1990, 42 U.S.C. 1396r-8 (c)(1)(C). This program requires drug manufacturers to provide rebates for medications dispensed to Medicaid patients. The amount of rebate is determined by a formula that takes into account the average manufacturer price (AMP) and the best price (or lowest price) offered to any other buyer; see also Ramsey Baghdadi, Medicaid Best Price, Health Affairs (Aug. 10, 2017), https://www.healthaffairs.org/do/10.1377/hpb20171008.000173 (“Program participation by drug manufacturers is essentially mandatory; companies declining to participate are excluded from all federal programs, including Medicare.”).

[61] The 340B Drug Pricing Program, established by the Veterans Health Care Act of 1992, requires drug manufacturers to provide outpatient drugs to eligible healthcare organizations and covered entities at significantly reduced prices. 42 U.S.C. § 256b (1993).

[62] Under the Affordable Care Act, a significant provision was introduced that directly affects the Medicare Part D coverage gap, commonly known as the “donut hole.” See 42 U.S.C. § 1395w-114a (2018). This provision mandates pharmaceutical manufacturers to offer a 50% discount on drugs for beneficiaries during this coverage gap. Id.

[63] See, e.g., Mark Duggan & Fiona M. Scott Morton, The Distortionary Effects of Government Procurement: Evidence from Medicaid Prescription Drug Purchasing (Nat’l Bureau of Econ. Res. Working Paper w10930, Feb. 2000), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=622874 (demonstrating that Medicaid pricing pressure on pharmaceuticals leads to downstream distortions in the price of pharmaceuticals purchased outside of the Medicaid program).

[64] Id.

[65] See, e.g., Sun et al., supra note 55. (discussing the fact that 90% of clinical trials fail, which means that the 10% of successful candidates effectively fund the experiments with the other 90%). As the authors note: Drug discovery and development is a long, costly, and high-risk process that takes over 10–15 years with an average cost of over $1–2 billion for each new drug to be approved for clinical use. For any pharmaceutical company or academic institution, it is a big achievement to advance a drug candidate to phase I clinical trial after drug candidates are rigorously optimized at preclinical stage. However, nine out of ten drug candidates after they have entered clinical studies would fail during phase I, II, III clinical trials and drug approval. It is also worth noting that the 90% failure rate is for the drug candidates that are already advanced to phase I clinical trial, which does not include the drug candidates in the preclinical stages. If drug candidates in the preclinical stage are also counted, the failure rate of drug discovery/development is even higher than 90%.

[66] John LaMattina, Pharma R&D Investments Moderating, But Still High, Forbes (Jun. 12, 2018), https://www.forbes.com/sites/johnlamattina/2018/06/12/pharma-rd-investments-moderating-but-still-high (Noting that R&D investment has typically been at 15% for the pharmaceutical industry).

[67] See Ekaterina Galkina Cleary, Matthew J. Jackson, Edward W. Zhou, & Fred D. Ledley, Comparison of Research Spending on New Drug Approvals by the National Institutes of Health vs the Pharmaceutical Industry, 2010-2019, 4(4) JAMA Health Forum (2023), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10148199. (“Funding from the NIH was contributed to 354 of 356 drugs (99.4%) approved from 2010 to 2019 totaling $187 billion, with a mean (SD) $1344.6 ($1433.1) million per target for basic research on drug targets and $51.8 ($96.8) million per drug for applied research on products.”)

[68] FDA, Step 3: Clinical Research (Jan. 4, 2018), https://www.fda.gov/patients/drug-development-process/step-3-clinical-research.

[69] Id.

[70] See, e.g., What’s Driving the Improvement in U.S. Cancer Survival Rates?, City of Hope (Jan. 26, 2023), https://www.cancercenter.com/community/blog/2023/01/cancer-survival-rates-are-improving Cancer death rates are down 33% since 1991. This is, in large part, due to the development of increasingly effective means of treating cancer and improving survivability odds.

[71] See supra notes 5-8 and accompanying text.

ICLE Director of Innovation Policy Kristian Stout joined fellow panelists Timothy B. Lee and Pamela Samuelson and moderator Brent Skorup to discuss the emerging legal issues surrounding artificial intelligence and its use of works protected under copyright law on a recent episode of the Federalist Society Regulatory Transparency Project’s Fourth Branch Podcast. The full episode is embedded below.

Here in New Jersey, where I live, the day before Halloween is commonly celebrated as “Mischief Night,” an evening of adolescent revelry and light vandalism that typically includes hurling copious quantities of eggs and toilet paper.

It is perhaps fitting, therefore, that President Joe Biden chose Oct. 30 to sign a sweeping executive order (EO) that could itself do quite a bit of mischief. And befitting the Halloween season, in proposing this broad oversight regime, the administration appears to be positively spooked by the development of artificial intelligence (AI).

The order, of course, embodies the emerging and now pervasive sense among policymakers that they should “do something” about AI; the EO goes so far as to declare that the administration feels “compelled” to act on AI. It largely directs various agencies to each determine how they should be involved in regulating AI, but some provisions go further than that. In particular, directives that set new reporting requirements—while ostensibly intended to forward the reasonable goal of transparency—could end up doing more harm than good.

Read the full piece here.

Introduction

We thank you for the opportunity to comment on this important notice of inquiry (NOI)[1] on artificial intelligence (AI) and copyright. We appreciate the U.S. Copyright Office undertaking a comprehensive review of the policy and copyright-law issues raised by recent advances in generative AI systems. This NOI covers key areas that require attention, from legal questions regarding infringement and fair use, to questions about how policy choices could shape opportunities for creators and AI producers to engage in licensing.

At this early date, AI systems have already generated some incredible visual art and impressive written texts, as well as a good deal of controversy. Some artists have banded together as part of an anti-AI campaign;^[2] lawsuits have been filed;^[3] and policy experts have attempted to think through the various legal questions raised by these machine-learning systems.

The debates over the role of AI in creative industries have particular salience for intellectual-property rights. Copyright is notoriously difficult to protect online, and the emergence of AI may exacerbate that difficulty. AI systems also potentially pose an additional wrinkle: it is at least arguable that the outputs they produce can themselves be considered unique creations. There are, of course, other open questions whose answers are relevant here, not the least being whether it is fair to assert that only a human can be “creative” (at least, so far).[4]

But leaving these questions aside, we can say that at least some AI systems produce unique outputs and are not merely routinely duplicating other pieces of work in a digital equivalent of collage. That is, at some level, the machines are engaged in a rudimentary sort of “learning” about how humans arrange creative inputs when generating images, music, or written works. The machines appear to be able to reconstruct this process and produce new sets of words, sounds, or lines and colors that conform to the patterns found in human art, in at least a simulacrum of “creativity.”

But that conclusion isn’t the end of the story. Even if some of these AI outputs are unique and noninfringing, the way that AI systems learn—by ingesting massive quantities of existing creative work—raises a number of thorny copyright-law issues. Indeed, some argue that these systems inherently infringe copyright during the learning phase and that, as discussed below, such processes may not survive a “fair use” analysis.

But nor is that assertion the end of the analysis. Rather, it raises the question of whether applying existing doctrine in this novel technological context yields the best results for society. Moreover, it heightens the need for a comprehensive analytical framework to help parse these questions.

A.            The Law & Economics of Copyright and AI

Nearly all would agree that it is crucial that law and public policy strike the appropriate balance between protecting creators’ existing rights and enabling society to enjoy the potentially significant benefits that could arise from the development of AI systems. Indeed, the subject is often cast as a dramatic conflict between creative professionals struggling to make ends meet and innovative firms working to provide cutting-edge AI technology. For the moment, however, it is likely more important to determine the right questions to ask and the proper analytical framework to employ than it is to identify any precise balancing point.

What is important to remember is that copyright policy is foremost economic in nature and “can be explained as a means for promoting efficient allocation of resources.”[5] That is to say, the reason that property rights in creative expression exist is to guarantee the continued production of such works.[6] The fundamental tradeoff in copyright policy is between the costs of limiting access to creative works, and the value obtained by encouraging production of such works.[7] The same applies in the context of AI: identifying the key tradeoffs and weighing the costs and benefits of restricting access to protected works by the producers (and users) of AI systems.[8]

This entails examining the costs and benefits of relatively stronger or weaker forms copyright protection in terms of their effects on both incentives and access, and as they relate to both copyright holders and AI-system developers. It also requires considering where the transaction costs should be allocated for negotiating access to both copyright and, as discussed infra,[9] the use of name/image/likeness, as well as how those allocations are likely to shape outcomes.

At root, these questions center on how to think about the property rights that limit access to protected works and, possibly even more importantly, how to assign new property rights governing the ability to control the use of a name/image/likeness. As we know from the work of the late Nobel laureate Ronald Coase, the actual demarcation of rights affects parties’ abilities to negotiate superior solutions.^[10] The development of nuisance law provides a good example of the problem at hand. When a legal regime provides either strict liability or no-liability rules around pollution, parties have little incentive to minimize harmful conduct:

The factory that has the absolute right to pollute will, if transaction costs are prohibitive, have no incentives to stop (or reduce) pollution even if the cost of stopping would be much less than the cost of pollution to the homeowners. Conversely, homeowners who have an absolute right to be free from pollution will, if transaction costs are prohibitive, have no incentive to take steps of their own to reduce the effects of pollution even if the cost to them of doing so (perhaps by moving away) is less than the cost to the factory of not polluting or of polluting less.^[11]

As Coase observed, this class of problem is best regarded as reciprocal in nature, and the allocation of rights matters in obtaining an efficient outcome. This is necessarily so because, when fully considered, B’s ability to restrain A from the pollution-generating activity can itself be conceived of as another kind of harm that B can impose on A. Therefore, the problem requires a balancing of the relative harms generated by both A and B in exercising conflicting claims in a particular context.

When thinking about how to minimize harms—whether from pollution or other activity that generates social costs (which is to say, nearly every activity)—the aim is to decide whether “the gain from preventing the harm is greater than the loss which would be suffered elsewhere as a result of stopping the action which produces the harm.”^[12] Theoretically, in a world without transaction costs, even assignments of no-liability or strict-liability rules could be bargained around. But we do not live in such a world.^[13] Thus, “[i]n a world in which there are costs of rearranging the rights established by the legal system [common law and statutory assignments of liability] are, in effect, making a decision on the economic problem and determining how resources are to be employed.”^[14]

While pollution rules, unlicensed uses of intellectual property, and a host of other activities subject to legal sanction are not typically framed as resource-allocation decisions, it is undeniable that they do have this character. This is true even where legislation attempts to correct deficiencies in the system. We experience a form of blindness when we focus on correcting what may be rightly perceived as problems in a liability regime. Such analysis tends to concentrate attention on particular deficiencies of the system and to nourish the belief that any measure that removes the deficiency is necessarily desirable. It diverts attention from other changes inevitably associated with the corrective measure—changes that may well produce more harm than the original deficiency.^[15]

All of this is to say that one solution to the costs generated by the need for AI systems to process a massive corpus of expensive, copyright-protected material is neither to undermine property rights, nor to make AI impossible, but to think about how new property rights could make the system work. It may be that some entirely different form or allocation of property right would facilitate bargaining between rightsholders and AI creators, optimizing resource allocation in a way the existing doctrinal regime may not be able to.

A number of other questions flow from this insight into the allocative nature of copyright. How would the incentives for human creators change under different copyright rules for AI systems, or in the face of additional rights? And how would access to copyrighted works for AI training change with different rules, and what effects would that access have on AI innovation?

Above all, our goal today should be to properly frame the AI and copyright debate by identifying tradeoffs, quantifying effects (where possible), and asking what rules best serve the overall objectives of the copyright system and the social goal of encouraging AI innovation. The best chance of striking the right balance will come from a rigorous framing of the questions and from the use of economic analysis to try to answer them.

B.            Copyright Law and AI: Moving Forward

As the Copyright Office undertakes this inquiry, it is important to recognize that, regardless of how the immediate legal questions around AI and copyright are resolved, the growing capabilities and adoption of generative AI systems will likely necessitate some changes in the long term.

The complex questions surrounding the intersection of AI and copyright law admit reasonable arguments on both sides. But AI is here to stay, regardless, and if copyright law is applied in an unduly restrictive manner that substantially hinders socially beneficial AI innovation, it could provoke a broader public-policy backlash that does more to harm copyright’s ability to protect creative works than it does to stanch AI’s ability to undermine it. Copyright law risks being perceived as an obstruction to technological progress if it is used preemptively to kill AI in the cradle. Such an outcome could galvanize calls for recalibrating copyright’s scope and protections in the name of the public interest.

This illustrates the precarious balancing act that copyright law faces in the wake of rapidly evolving technologies like AI. Aggressive copyright restrictions that curtail AI development could instigate a public-policy counter-reaction before Congress and the courts that ultimately undermines copyright’s objectives. The judicious course is to adapt copyright law cautiously to enable AI’s responsible evolution, while resolutely preserving the incentives for human creativity.

In the remainder of this analysis, we offer our perspective on the likely outcomes of the AI-copyright issues raised in this NOI, given the current state of the law. These assessments reflect our perspective formed through the rigorous application of established copyright principles and precedent to the novel technological context of generative AI systems. Reasonable arguments rooted in existing doctrine could be made to support different conclusions. We submit these comments not as definitive predictions or normative preferences, but rather as informed appraisals of how courts may analyze AI under present copyright law, absent legislative intervention.

We appreciate the Copyright Office starting this process to modernize copyright law for the AI age. This inquiry is an important first step, but openness to further evolution will be key to promoting progress in both AI and the arts. We believe an open, evidence-based discussion of these issues will lead to balanced solutions that uphold copyright’s constitutionally mandated purpose, while allowing responsible AI innovation for the public benefit.

II.            The Training of AI Systems and the Applicability of Fair Use

In the NOI, the Copyright Offices asks: “[u]nder what circumstances would the unauthorized use of copyrighted works to train AI models constitute fair use?”^[16]

To answer this question, it would be useful to first briefly walk through a high-level example of how AI systems work, in order to address the most relevant points of contact between AI systems and copyright law.

A.            A Brief Technical Description of AI Training

AI-generated content is not a single “thing,” but a collection of differing processes, each with different implications for the law. For the purposes of this discussion, we will discuss image generation using “generated adversarial networks” (GANs) and diffusion models. Although different systems and different types of content generation will vary, the basic concepts discussed below are nonetheless useful at a general level.^[17]

A GAN is a type of machine-learning model that consists of two parts: a generator and a discriminator.^[18] The generator is trained to create new images that look like they come from a particular dataset, while the discriminator is trained to distinguish the generated images from real images in its original dataset.^[19] The two parts are trained together in an adversarial manner, with the generator trying to produce images that can fool the discriminator and the discriminator trying to correctly identify the generated images.^[20]

A diffusion model, by contrast, analyzes the distribution of information in an image, as noise is progressively added to it.^[21] This kind of algorithm analyzes characteristics of sample images, like the distribution of colors or lines, in order to understand what counts as an accurate representation of a subject (i.e., what makes a picture of a cat look like a cat, and not like a dog).^[22]

For example, in the generation phase, diffusion-based systems start with randomly generated noise, and work backward in “denoising” steps to essentially “see” shapes:

The sampled noise is predicted so that if we subtract it from the image, we get an image that’s closer to the images the model was trained on (not the exact images themselves, but the distribution – the world of pixel arrangements where the sky is usually blue and above the ground, people have two eyes, cats look a certain way – pointy ears and clearly unimpressed).^[23]

While it is possible that some implementations might be designed in a way that saves copies of the training images,^[24] for at least some systems, once the network is trained using these techniques, it will not need to rely on saved copies of input work in order to produce outputs. The models that are produced during training are, in essence, instructions to a different piece of software about how to start with a prompt from a user, a palette of pure noise, and progressively “discover” signal in that image until some new image emerges.

B.            Fair Use

The creator of some of the most popular AI tools, OpenAI, is not shy about their use of protected works in the training phase of the algorithms. In comments to the U.S. Patent and Trademark Office (PTO), OpenAI noted that:

Modern AI systems require large amounts of data. For certain tasks, that data is derived from existing publicly accessible “corpora”… of data that include copyrighted works. By analyzing large corpora (which necessarily involves first making copies of the data to be analyzed), AI systems can learn patterns inherent in human-generated data and then use those patterns to synthesize similar data which yield increasingly compelling novel media in modalities as diverse as text, image, and audio. (emphasis added).^[25]

Thus, at the training stage, the most popular forms of AI systems require making copies of existing works. And where that material is either not in the public domain or is not licensed, an infringement can occur. Thus, the copy must not be infringing (say, because it is transient), or some affirmative defense is needed to excuse the infringement. Toward this end, OpenAI believes that this use should qualify as fair use,^[26] as do most or all the other major producers of generative AI systems.^[27]

But as OpenAI has framed the fair-use analysis, it is not clear that these uses should qualify. There are two major questions in this respect: will the data used to train these systems count as “copies” under the Copyright Act, and, if so, is the use of these “copies” sufficiently “transformative” to qualify for the fair-use defense?

1.              Are AI systems being trained with ‘copies’ of protected works?

Section 106 of the Copyright Act grants the owner of a copyright the exclusive right “to reproduce… copyrighted work in copies” and to authorize others to do so.^[28] If an AI system makes a copy of a file to a computer during training, this would likely constitute a prima facie violation of the copyright owner’s exclusive right of reproduction under Section 106. This is fairly straightforward.

But what if the “copy” is “transient” and/or only partial pieces of content are used in the training? For example, what if a training program merely streamed small bits of a protected work into temporary memory as part of its training, and retained no permanent copy?

As the Copyright Office has previously observed, even temporary reproductions of a work in a computer’s memory can constitute “copies” under the Copyright Act.^[29] Critically, this includes even temporary reproductions made as part of a packet-switching network transmission, where a particular file is broken into individual packets, because the packets can be reassembled into substantial portions or even entire works.^[30] On the topic of network-based transmission, the Copyright Office further observed that:

Digital networks permit a single disk copy of a work to meet the demands of many users by creating multiple RAM copies. These copies need exist only long enough to be perceived (e.g., displayed on the screen or played through speakers), reproduced or otherwise communicated (e.g., to a computer’s processing unit) in order for their economic value to be realized. If the network is sufficiently reliable, users have no need to retain copies of the material. Commercial exploitation in a network environment can be said to be based on selling a right to perceive temporary reproductions of works.^[31]

This is a critical insight that translates well to the context of AI training. The “transience” of the copy matters with respect to the receiver’s ability to perceive the work in a way that yields commercial value. Under this reasoning, the relevant locus of analysis is on the AI system’s ability to “perceive” a work for the purposes of being trained to “understand” the work. In this sense, you could theoretically find the existence of even more temporary copies than that necessary for human perception to implicate the reproduction right.

Even where courts have been skeptical of extending the definition of “copy” to “fleeting” copies in computer memory, this underlying logic is revealed. In Cartoon Network LP, LLLP v. CSC Holdings, Inc., 536 F.3d 121 (2008), the 2nd U.S. Circuit Court of Appeals had to determine whether buffered media sent to a DVR device was too “transient” to count as a “copy”:

No bit of data remains in any buffer for more than a fleeting 1.2 seconds. And unlike the data in cases like MAI Systems, which remained embodied in the computer’s RAM memory until the user turned the computer off, each bit of data here is rapidly and automatically overwritten as soon as it is processed. While our inquiry is necessarily fact-specific, and other factors not present here may alter the duration analysis significantly, these facts strongly suggest that the works in this case are embodied in the buffer for only a “transitory” period, thus failing the duration requirement.^[32]

In Cartoon Network, the court acknowledged both that the duration analysis was fact-bound, and also that the “fleeting” nature of the reproduction was important. “Fleeting” is a relative term, based on the receiver’s capacities. A ball flying through the air may look “fleeting” to a human observer, but may appear to go much more cognizable to a creature with faster reaction time, such as a house fly. So, too, with copies of a work in a computer’s memory and the ability to “perceive” what is fixed in a buffer: what may be much too quick for a human to perceive may very well be within an AI system’s perceptual capabilities.

Therefore, however the training copies are held, there is a strong possibility that a court will find them to be “copies” for the purposes of the reproduction right—even with respect to partial copies that exist for very small amounts of time.

2.              The purpose and character of using protected works to train AI systems

Fair use provides for an affirmative defense against infringement when the use is, among other things, “for purposes such as criticism, comment, news reporting, teaching…, scholarship, or research.”^[33] When deciding whether a fair-use defense is applicable, a court must balance a number of factors:

1. the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
2. the nature of the copyrighted work;
3. the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
4. the effect of the use upon the potential market for or value of the copyrighted work.^[34]

The fair-use defense that AI creators have advanced is rooted in the first factor: the nature and character of the use. Although a full analysis of all the factors is ultimately necessary, analysis of the first factor is sufficiently complicated to warrant full attention here. In particular, the complex issue at hand is whether uses of protected works to train AI systems are sufficiently “transformative” or not.^[35]

Whether the use of a copyrighted work to train an AI is “transformative” is certainly a novel question, but it is one that will likely be answered in light of an observation the U.S. Supreme Court made in Campbell v. Acuff Rose Music:

[W]hen a commercial use amounts to mere duplication of the entirety of an original, it clearly “supersede[s] the objects,”… of the original and serves as a market replacement for it, making it likely that cognizable market harm to the original will occur… But when, on the contrary, the second use is transformative, market substitution is at least less certain, and market harm may not be so readily inferred.^[36]

Moreover, “[t]he word ‘transformative’ cannot be taken too literally as a sufficient key to understanding the elements of fair use. It is rather a suggestive symbol for a complex thought, and does not mean that any and all changes made to an author’s original text will necessarily support a finding of fair use.”[37] A key question, then, is whether training AI systems on copyrighted works amounts to a mere “duplication of the entirety of an original” or is sufficiently “transformative” to support a fair-use defense. As noted above, OpenAI believes that its use is transformative. According to its comments:

Training of AI systems is clearly highly transformative. Works in training corpora were meant primarily for human consumption for their standalone entertainment value. The “object of the original creation,” in other words, is direct human consumption of the author’s ?expression.? Intermediate copying of works in training AI systems is, by contrast, “non-expressive” the copying helps computer programs learn the patterns inherent in human-generated media. The aim of this process—creation of a useful generative AI system—is quite different than the original object of human consumption. The output is different too: nobody looking to read a specific webpage contained in the corpus used to train an AI system can do so by studying the AI system or its outputs. The new purpose and expression are thus both highly transformative.^[38]

This framing, however, works against OpenAI’s interests. As noted above, and reinforced in the immediately preceding quote, generative AI systems are made of at least two distinct pieces. The first is a piece of software that ingests existing works and creates a file that can serve as instructions to the second piece of software. The second piece of software takes the output of the first and can produce independent results. Thus, there is a clear discontinuity in the process whereby the ultimate work created by the system is disconnected from the creative inputs used to train the software.

Therefore, the protected works are arguably ingested into the first part of the system “for their standalone entertainment value.” That is to say, the goal of copying and showing a protected work to an AI system is for the analog of “direct human consumption of the author’s expression” in order for the system to learn about that expression.

The software is learning what counts as “standalone entertainment value” and therefore the works must be used in those terms. Surely, a computer is not sitting on a couch and surfing for its pleasure. But it is solely for the very “standalone entertainment value” that the first piece of software is being shown copyrighted material. By contrast, parody or “remixing” uses incorporate a work into some secondary expression that directly transforms the input. The way these systems work is to learn what makes a piece entertaining and then to discard that piece altogether. Moreover, this use for the art qua art most certainly interferes with the existing market, insofar as this use is in lieu of reaching a licensing agreement with rightsholders.

A good analogy is art students and art textbooks. Art students view protected works in an art textbook in order to learn how to reproduce the styles contained therein. The students would not be forgiven for pirating the textbooks merely because they intend to go on to make new paintings. They would still be liable for copyright infringement if they used unlicensed protected works as part of their education.

The 2nd U.S. Circuit Court of Appeals dealt with a case that operates similarly to this dynamic. In American Geophysical Union v. Texaco, 60 F.3d 913 (2d Cir. 1994), the 2nd Circuit considered whether Texaco’s photocopying of scientific articles produced by the plaintiffs qualified for a fair-use defense. Texaco employed between 400 and 500 research scientists and, as part of supporting their work, maintained subscriptions to a number of scientific journals.^[39]

It was common practice for Texaco’s scientists to photocopy entire articles and save them in a file.^[40] The plaintiffs sued for copyright infringement.^[41] Texaco asserted that photocopying by its scientists for the purposes of furthering scientific research—that is to train the scientists on the content of the journal articles—should count as a fair use. The argument was, at least in part, that this was sufficiently “transformative,” because the scientists were using that knowledge to invent new products.^[42] The 2nd Circuit disagreed:

The “transformative use” concept is pertinent to a court’s investigation under the first factor because it assesses the value generated by the secondary use and the means by which such value is generated. To the extent that the secondary use involves merely an untransformed duplication, the value generated by the secondary use is little or nothing more than the value that inheres in the original. Rather than making some contribution of new intellectual value and thereby fostering the advancement of the arts and sciences, an untransformed copy is likely to be used simply for the same intrinsic purpose as the original, thereby providing limited justification for a finding of fair use….^[43]

The 2nd Circuit thus observed that copies of the scientific articles were made solely to consume the material itself. AI developers often make an argument analogous to that made by Texaco: that training AI systems surely advances scientific research, and therefore fosters the “advancement of the arts and sciences.” But in American Geophysical Union, the initial copying of copyrighted content, even where it was ultimately used for the “advancement of the arts and sciences,” was not held to be sufficiently “transformative.”^[44] The case thus stands for the proposition that one cannot merely identify a social goal down that would be advanced at some future date in order to permit an exception to copyright protection. As the court put it:

[T]he dominant purpose of the use is a systematic institutional policy of multiplying the available number of copies of pertinent copyrighted articles by circulating the journals among employed scientists for them to make copies, thereby serving the same purpose for which additional subscriptions are normally sold, or… for which photocopying licenses may be obtained.[45]

The use itself must be transformative and different, and copying is not transformative merely because it may be used as an input into a later transformative use. By the same token, therefore, it seems likely that where an AI system ingests (copies) copyrighted works, that use is similarly not transformative, despite its ultimate use as an input in the creation of other original works.

Comparing the American Geophysical Union analysis with the search-engine “snippets” and “thumbnails” cases provides a useful comparison relevant to the AI analysis. In Kelly v. Arriba Soft Corp., 336 F.3d 811 (9^th Cir. 2002), the 9^th U.S. Circuit Court of Appeals ruled that a search engine’s creation of thumbnail images from original copies was a transformative fair use.[46] Arriba’s search-engine crawler made full-sized copies of Kelly’s images and stored them temporarily on Arriba’s server to generate thumbnail versions. After the thumbnails were created, the full-sized originals were deleted. The thumbnails were used to facilitate Arriba’s image-based search engine. In reaching its fair-use conclusion, the 9^th Circuit opined that:

Arriba’s use of Kelly’s images promotes the goals of the Copyright Act and the fair use exception. The thumbnails do not stifle artistic creativity because they are not used for illustrative or artistic purposes and therefore do not supplant the need for the originals.[47]

Further, although “Arriba made exact replications of Kelly’s images, the thumbnails were much smaller, lower-resolution images that served an entirely different function than Kelly’s original images.”[48]

The court found it important that the search engine did not use the protected works for their intended “aesthetic experience,” but rather for the purpose of constructing a search index.[49] Indeed, the entire point of a search engine is not to “supersede” the original, but in many or most cases to provider users an efficient means to find that original online.[50]

The court discussed, but only briefly, the benefit to the public of Arriba’s transformative use,[51] noting that “[Arriba’s thumbnails] benefit the public by enhancing information-gathering techniques on the internet.”[52] Five years later, in Perfect 10 Inc. v. Amazon.com Inc., 487 F.3d 701 (2007), the 9^th Circuit expanded on this question somewhat.[53] There, in holding that the novelty of the use was of crucial importance to the analysis,[54] the court also stressed that the value of that use was a function of its newness:

[A] search engine provides social benefit by incorporating an original work into a new work, namely, an electronic reference tool. Indeed, a search engine may be more transformative than a parody [the use at issue in Campbell] because a search engine provides an entirely new use for the original work, while a parody typically has the same entertainment purpose as the original work.[55]

Indeed, even in light of the commercial nature of Google’s use of copyrighted content in its search engine, its significant public benefit carried the day: “We conclude that the significantly transformative nature of Google’s search engine, particularly in light of its public benefit, outweighs Google’s superseding and commercial uses of the thumbnails in this case.”[56] And, of particular relevance to these questions in the context of AI, the court in Perfect 10 went on to “note the importance of analyzing fair use flexibly in light of new circumstances.”[57]

Ultimately, the Perfect 10 decision tracked Kelly fairly closely on the rest of the “transformativeness” analysis in finding fair use, because “[a]lthough an image may have been created originally to serve an entertainment, aesthetic, or informative function, a search engine transforms the image into a pointer directing a user to a source of information.”[58]

The core throughline in this line of cases is the question of whether a piece of content is being used for its expressive content, weighed against the backdrop of whether the use is for some new (and, thus, presumptively valuable) purpose. In Perfect 10 and Kelly, the transformative use was the creation of a search index.

“Snippets” fair-use cases track a similar line of reasoning. For example, in Authors Guild v. Google Inc., 804 F.3d 202 (2d Cir. 2015), the 2nd Circuit ruled that Google’s use of “snippets” of copyrighted books in its Library Project and Google Books website was a “transformative” fair use.[59] Holding that the “snippet view” of books digitized as part of the Google Books project did not constitute an effectively competing substitute to the original works, the circuit court noted that copying for the purpose of “criticism” or—as in that case—copying for the purpose of “provision of information about” the protected work, “tends most clearly to satisfy Campbell’s notion of the ‘transformative’ purpose.”[60]

Importantly, the court emphasized the importance of the public-benefit aspect of transformative uses: “[T]ransformative uses tend to favor a fair use finding because a transformative use is one that communicates something new and different from the original or expands its utility, thus serving copyright’s overall objective of contributing to public knowledge.”[61]

Underscoring the idea that the “transformativeness” analysis weighs whether a use is merely for expressive content against the novelty/utility of the intended use, the court observed:

Google’s division of the page into tiny snippets is designed to show the searcher just enough context surrounding the searched term to help her evaluate whether the book falls within the scope of her interest (without revealing so much as to threaten the author’s copyright interests). Snippet view thus adds importantly to the highly transformative purpose of identifying books of interest to the searcher.[62]

Thus, the absence of use of the work’s expressive content, coupled with a fairly circumscribed (but highly novel) use was critical to the outcome.

The entwined questions of transformative use and the public benefit it confers are significantly more complicated in the AI context, however. Unlike the incidental copying involved in search-engine indexing or thumbnails, training generative AI systems directly leverages copyrighted works for their expressive value. In the Google Books and Kelly cases, the defendant systems extracted limited portions of works or down-sampled images solely to identify and catalog their location for search purposes. The copies enabled indexing and access, and they expanded public knowledge through a means unrelated to the works’ protected aesthetics.

But in training AI models on copyrighted data, the systems necessarily parse the intrinsic creative expression of those works. The AI engages with the protected aesthetic elements themselves, not just superficial markers (like title, length, location on the internet, etc.), in order to internalize stylistic and compositional principles. This appropriates the heart of the works’ copyright protection for expressive ends, unlike the more tenuous connections in search systems.

The AI is thus “learning” directly from the protected expression in a manner akin to a human student studying an art textbook, or like the scientists learning from the journals in American Geophysical Union. The subsequent AI generations are built from mastery of the copyrighted training materials’ creative expression. Thus, while search-engine copies only incidentally interact with protected expression to enable unrelated innovation, AI training is predicated on excavating the protected expression itself to fuel iterative creation. These meaningfully different purposes have significant fair-use implications.

This functional difference is, as noted, central to the analysis of a use’s “purpose and character.” Indeed, “even making an exact copy of a work may be transformative so long as the copy serves a different function than the original work.”[63] But the benefit to the public from the new use is important, as well, particularly with respect to the possible legislative response that a restrictive interpretation of existing doctrine may engender.

If existing fair-use principles prohibit the copying required for AI, absent costly item-by-item negotiation and licensing, the transaction costs could become prohibitive, thwarting the development of technologies that promise great public value.[64] Copyright law has faced similar dilemmas before, where the transaction costs of obtaining permission for socially beneficial uses could frustrate those uses entirely.[65] In such cases, we have developed mechanisms like compulsory licensing to facilitate the necessary copying, while still attempting to compensate rightsholders. An unduly narrow fair-use finding for AI training could spur calls for similar interventions in service of enabling AI progress.

In other words, regardless of the veracity of the above conclusion that AI’s use of copyrighted works may not, in fact, serve a different function than the original, courts and legislators may be reluctant to allow copyright doctrine to serve as an absolute bar against self-evidently valuable activity like AI development. Our aim should be to interpret or recalibrate copyright law to permit such progress while upholding critical incentives for creators.

C.            Opt-In vs. Opt-Out Use of Protected Works

The question at the heart of the prior discussion—and, indeed, at the heart of the economic analysis of copyright—is whether the transaction costs that accompany requiring express ex ante permission for the use of protected works are so high that they impedes socially beneficial conduct whose value would outweigh the social cost of allowing permissionless and/or uncompensated use.[66] The NOI alludes to this question when it asks: “Should copyright owners have to affirmatively consent (opt in) to the use of their works for training materials, or should they be provided with the means to object (opt out)?”^[67]

This is a complex problem. Given the foregoing thoughts on fair use, it seems quite possible that, at present, the law requires creators of AI systems to seek licenses for protected content, or else must resort to public-domain works for training. Given the volume of copyrighted works that AI developers currently use to train these systems, such requirements may be broadly infeasible.

On one hand, requiring affirmative opt-in consent from copyright holders imposes significant transaction costs on AI-system developers to identify and negotiate licenses for the vast amounts of training data required. This could hamper innovation in socially beneficial AI systems. On the other hand, an opt-out approach shifts more of the transaction-cost burden to copyright holders, who must monitor and object to unwanted uses of their works. This raises concerns about uncompensated use.

Ultimately, the question is where the burden should lie: with AI-system developers to obtain express consent, or with copyright holders to monitor and object to uses? Requiring some form of consent may be necessary to respect copyright interests. Yet an opt-out approach may strike the right balance, by shifting some of the burden back to AI developers while avoiding the infeasibly high transaction costs of mandatory opt-in consent. The optimal approach likely involves nuanced policymaking to balance these competing considerations. Moreover, as we discuss infra, the realistic outcome is most likely going to require rethinking the allocation of property rights in ways that provide for large-scale licensing. Ideally, this could be done through collective negotiation, but perhaps at a de minimis rate, while allowing creators to bargain for remuneration on the basis of other rights, like a right of publicity or other rights attached to the output of AI systems, rather than the inputs.[68]

1.              Creator consent

Relatedly, the Copyright Office asks: “If copyright owners’ consent is required to train generative AI models, how can or should licenses be obtained?”^[69]

Licensing markets exist, and it is entirely possible that major AI developers and large groups of rightsholders can come to mutually beneficial terms that permit a sufficiently large body of protected works to be made available as training data. Something like a licensing agency for creators who choose to make their works available could arise, similar to the services that exist to provide licensed music and footage for video creators.^[70] It is also possible for some to form collective-licensing organizations to negotiate blanket permissions covering many works.

It’s important to remember that our current thinking is constrained by our past experience. All we know today are AI models trained on vast amounts of unlicensed works. It is entirely possible that, if firms were required to seek licenses, unexpected business models would emerge to satisfy both sides of the equation.

For example, an AI firm could develop its own version of YouTube’s ContentID, which would allow creators to control when their work is used in AI training. For some well-known artists, this could be negotiated with an upfront licensing fee. On the user side, any artist who has opted in could then be selected as a “style” for the AI to emulate—triggering a royalty payment to the artist when a user generates an image or song in that style. Creators could also have the option of removing their influence from the system if they so desire.

Undoubtedly, there are other ways to structure the relationship between creators and AI systems  that would facilitate creators’ monetization of the use of their work in AI systems, including legal and commercial structures that create opportunities for both creators and AI firms to succeed.

III.          Generative AI Outputs: Protection of Outputs and Outputs that Infringe

The Copyright Office asks: “Under copyright law, are there circumstances when a human using a generative AI system should be considered the ‘author’ of material produced by the system?”^[71]

Generally speaking, we see no reason why copyright law should be altered to afford protection to purely automatic creations generated by AI systems. That said, when a human makes a nontrivial contribution to generative AI output—such as editing, reframing, or embedding the AI-generated component within a larger work—the resulting work should qualify for copyright protection.

Copyright law centers on the concept of original human authorship.^[72] The U.S. Constitution expressly limits copyright to “authors.”^[73] As of this writing, however, generative AI’s capacities do not rise to the level of true independent authorship. AI systems remain tools that require human direction and judgment.[74] As such, when a person provides the initial prompt or framing, makes choices regarding the iterative development of the AI output, and decides that the result is satisfactory for inclusion in a final work, they are fundamentally engaging in creative decision making that constitutes authorship under copyright law.

As Joshua Gans has observed of recent Copyright Review Board decisions:

Trying to draw some line between AI and humans with the current technology opens up a massive can of worms. There is literally no piece of digital work these days that does not have some AI element to it, and some of these mix and blur the lines in terms of what is creative and what is not. Here are some examples:

A music artist uses AI to denoise a track or to add an instrument or beat to a track or to just get a composition started.

A photographer uses Photoshop or takes pictures with an iPhone that already uses AI to focus the image and to sort a burst of images into one that is appropriate.

A writer uses AI to prompt for some dialogue when stuck at some point or to suggest a frame for writing a story.[75]

Attempting to separate out an “AI portion” from the final work, as the Copyright Review Board proposed, fundamentally misunderstands the integrated nature of the human-AI collaborative process. The AI system cannot function without human input, and its output remains raw material requiring human creativity to incorporate meaningfully into a finished product.

Therefore, when a generative AI system is used as part of a process guided by human creative choices, the final work should be protected by copyright, just as a work created using any other artistic tool or collaborator would be. Attenuating copyrightability due to the use of AI would undermine basic copyright principles and fail to recognize the essentially human nature of the creative process.

A.            AI Outputs and Infringement

The NOI asks: “Is the substantial similarity test adequate to address claims of infringement based on outputs from a generative AI system, or is some other standard appropriate or necessary?” (Question 23)

The outputs of AI systems may or may not violate IP laws, but there is nothing inherent in the processes described above that dictates that they must. As noted, the most common AI systems do not save copies of existing works, but merely “instructions” (more or less) on how to create new work that conforms to patterns found by examining existing work. If we assume that a system isn’t violating copyright at the input stage, it’s entirely possible that it can produce completely new pieces of art that have never before existed and do not violate copyright.

They can, however, be made to violate copyrights. For example, these systems can be instructed to generate art, not just in the style of a particular artist, but art that very closely resembles existing pieces. In this sense, it would be making a copy that theoretically infringes. The fact of an AI’s involvement would not change the analysis: just as with a human-created work, if it is substantially similar to a copyrighted work, it may be found infringing.

There is, however, a common bug in AI systems that leads to outputs that are more likely to violate copyright in this way. Known as “overfitting,” the training leg of these AI systems can be presented with samples that contain too many instances of a particular image.^[76] This leads to a dataset that contains too much information about the specific image, such that—when the AI generates a new image—it is constrained to producing something very close to the original. Similarly, there is evidence that some AI systems are “memorizing” parts of protected books.^[77] This could lead to AI systems repeating copyright-protected written works.

1.              The substantial-similarity test

The substantial-similarity test remains functionally the same when evaluating works generated using AI. To find “substantial similarity,” courts require evidence of copying, as well as an expression that is substantially similar to a protected work.^[78] “It is now an axiom of copyright law that actionable copying can be inferred from the defendant’s access to the copyrighted work and substantial similarity between the copyrighted work and the alleged infringement.”^[79] In many or most cases, it will arguably be the case that AI systems have access to quite a wide array of protected works that are posted online. Thus, there may not be a particularly high hurdle to determine that an AI system actually copied a protected work.

There is, however, one potential problem for the first prong of this analysis. Models produced during a system’s training process do not (usually) contain the original work, but are the “ideas” that the AI systems generated during training. Thus, where the provenance of works contained in a training corpus is difficult to source, it may not be so straightforward to make inferences about whether a model “saw” a particular work. This is because the “ideas” that the AI “learns” from its training corpus are unprotected under U.S. copyright law, as it is permissible to mimic unprotected elements of a copyrighted work (such as ideas).^[80]

Imagine a generative AI system trained on horror fiction. It would be possible for this system to produce a new short story that is similar to one written by Stephen King, but the latent data in the model almost certainly would not violate any copyrights that King holds in his work. The model would contain “ideas” about horror stories, including those learned from an array of authors who were themselves influences on Stephen King, and potentially some of King’s own stories. What the AI system “learns” in this case is the relationship between words and other linguistic particularities that are commonly contained in horror fiction. That is, it has “ideas” about what goes into a horror story, not (theoretically) the text of the horror story itself.

Thus, when demonstrating indirect proof of copying in the case of a Stephen King story, it may pose a difficulty that an AI system has ingested all of H.P. Lovecraft’s work—an author who had a major influence on King. The “ideas” in the model and the output it subsequently produces may, in fact, produce something similar to a Stephen King work, but it may have been constructed largely or entirely on material from Lovecraft and other public-domain horror writers. The problem becomes only more complicated when you realize that this system could also have been trained on public-domain fan fiction written in the style of Stephen King. Thus, for the purposes of the first prong of this analysis, courts may place greater burden on plaintiffs in copyright actions against model producers to demonstrate more than merely that a work was merely available online.

Assuming that plaintiffs are able to satisfy the first prong, once an AI system “expresses” those ideas, that expression could violate copyright law under the second prong of the substantial-similarity test. The second prong inquires whether the final work appropriated the protected original expression.^[81] Any similarities in unprotectable ideas, facts, or common tropes are disregarded.^[82] So, in both traditional and AI contexts, the substantial-similarity test ultimately focuses on the protected components of creative expression, not surface similarity.

The key determination is whether the original work’s protected expression itself has been impermissibly copied, no matter the process that generated the copy. AI is properly viewed as simply another potential tool that could be used in certain acts of copying. It does not require revisiting settled principles of copyright law.

B.            Direct and Secondary Liability

The NOI asks: “If AI-generated material is found to infringe a copyrighted work, who should be directly or secondarily liable—the developer of a generative AI model, the developer of the system incorporating that model, end users of the system, or other parties?”[83]

Applying traditional copyright-infringement frameworks to AI-generated works poses unique challenges in determining direct versus secondary liability. In some cases, the AI system itself may create infringing content without any direct human causation.

1.              Direct liability

If the end user prompts an AI system in a way that intentionally targets copyrighted source material, they may meet the threshold for direct infringement by causing the AI to reproduce protected expression.^[84] Though many AI prompts contain only unprotected ideas, users may sometimes input copyrightable material as the basis for the AI output. For example, a user could upload a copyrighted image and request the AI to make a new drawing based on the sample. In such cases, the user is intentionally targeting copyrighted works and directly “causing” the AI system to reproduce output that is similar. If sufficiently similar, that output could infringe on the protected input. This would be a question of first impression, but it is a plausible reading of available cases.

For example, in CoStar Grp. Inc. v. LoopNet Inc., 373 F.3d 544 (4th Cir. 2004), the 4^th U.S. Circuit Court of Appeals had to consider whether an internet service provider (ISP) could be directly liable when third parties reposted copyrighted material owned by the plaintiff. In determining that merely owning the “machine” through which copies were made or transmitted was not enough to “cause” a direct infringement, the court held that:

[T]o establish direct liability under §§ 501 and 106 of the Act, something more must be shown than mere ownership of a machine used by others to make illegal copies. There must be actual infringing conduct with a nexus sufficiently close and causal to the illegal copying that one could conclude that the machine owner himself trespassed on the exclusive domain of the copyright owner. The Netcom court described this nexus as requiring some aspect of volition or causation… Indeed, counsel for both parties agreed at oral argument that a copy machine owner who makes the machine available to the public to use for copying is not, without more, strictly liable under § 106 for illegal copying by a customer. The ISP in this case is an analogue to the owner of a traditional copying machine whose customers pay a fixed amount per copy and operate the machine themselves to make copies. When a customer duplicates an infringing work, the owner of the copy machine is not considered a direct infringer. Similarly, an ISP who owns an electronic facility that responds automatically to users’ input is not a direct infringer.^[85]

Implied in the 4^th Circuit’s analogy is that, while the owner of a copying machine might not be a direct infringer, a user employing such a machine could be a direct infringer. It’s an imperfect analogy, but a user of an AI system prompting it to create a “substantially similar” reproduction of a protected work could very well be a direct infringer under this framing. Nevertheless, the analogy is inexact, because the user feeds an original into a copying machine in order to make a more-or-less perfect copy of the original, whereas an AI system generates something new but similar. The basic mechanism of using a machine to try to reproduce a protected work, however, remains essentially the same. Whether there is an infringement would be a question of “substantial similarity.”

2.              Secondary liability

As in the case of direct liability, the nature of generative AI makes the secondary-liability determination slightly more complicated, as well. That is, paradoxically, the basis for secondary liability could theoretically arise even where there was no direct infringement.[86]

The first piece of this analysis is relatively easier. If a user is directly liable for infringing a protected work, as noted above, the developer and provider of a generative AI system may face secondary copyright liability. If the AI developer or distributor knows the system can produce infringing outputs, and provides tools or material support that allows users to infringe, it may be liable for contributory infringement.^[87] Critically, merely designing a system that is capable of infringing is not enough to find contributory liability.^[88]

An AI producer or distributor may also have vicarious liability, insofar as it has the right and ability to supervise users’ activity and a direct financial interest in that activity.^[89] AI producers have already demonstrated their ability to control users’ behavior to thwart unwanted uses of the service.^[90] Thus, if there is a direct infringement by a user, a plausible claim for vicarious liability could be made so long as there is sufficient connection between the user’s behavior and the producer’s financial interests.

The question becomes more complicated when a user did not direct the AI system to infringe. When the AI generates infringing content without user direction, it’s not immediately clear who would be liable for the infringement.^[91] Consider the case where, unprompted by either the user or the AI producer, an AI system creates an output that would infringe under the substantial-similarity test. Assuming that the model has not been directed by the producer to “memorize” the works it ingests, the model itself consists of statistical information about the relationship between different kinds of data. The infringer, in a literal sense, is the AI system itself, as it is the creator of the offending output. Technically, this may be a case of vicarious liability, even without an independent human agent causing the direct infringement.

We know that copyright protection can only be granted to humans. As the Copyright Review Board recently found in a case deciding whether AI-generated outputs can be copyrighted:

The Copyright Act protects, and the Office registers, “original works of authorship fixed in any tangible medium of expression.” 17 U.S.C. § 102(a). Courts have interpreted the statutory phrase “works of authorship” to require human creation of the work.^[92]

But can an AI system directly violate copyright? In his Aereo dissent, Justice Clarence Thomas asserted that it was a longstanding feature of copyright law that violation of the performance right required volitional behavior.^[93] But the majority disagreed with him, holding that, by running a fully automated system of antennas intended to allow users to view video at home, the system gave rise to direct copyright liability.^[94] Thus, implied in the majority’s opinion is the idea that direct copyright infringement does not require “volitional” conduct.

It is therefore plausible that a non-sentient, fully automated AI system could infringe copyright, even if, ultimately, there is no way to recover against the nonhuman agent. That does, however, provide an opportunity for claims of vicarious liability against the AI producer or distributor— at least, where the producer has the power to control the AI system’s behavior and that behavior appears to align with the producer’s financial interests.

3.              Protecting the ‘style’ of human creators

The NOI asks: “Are there or should there be protections against an AI system generating outputs that imitate the artistic style of a human creator (such as an AI system producing visual works ‘in the style of’ a specific artist)?”[95]

At the federal level, one candidate for protection against AI imitating some aspects of a creator’s works can currently be found in trademark law. Trademark law, governed by the Lanham Act, protects names, symbols, and other source identifiers that distinguish goods and services in commerce.^[96] Unfortunately, a photograph or likeness, on its own, typically does not qualify for trademark protection, unless it is consistently used on specific goods.^[97] Even where there is a likeness (or similar “mark”) used consistently as part of branding a distinct product, many trademark-infringement claims would be difficult to establish in this context, because trademark law does little to protect many aspects of a creator’s work.

Moreover, the Supreme Court has been wary about creating a sort of “mutant copyright” in cases that invoke the Lanham Act as a means to enforce a sort of “right of attribution,” which would potentially give creators the ability to control the use of their name in broader contexts.^[98] In this context, the Court has held that the relevant parts of the Lanham Act were not designed to “protect originality or creativity,”^[99] but are focused solely on “actions like trademark infringement that deceive consumers and impair a producer’s goodwill.”^[100]

In many ways, there is a parallel here to the trademark cases involving keyword bidding in online ads. At a high level, search engines and other digital-advertising services do not generally infringe trademark when they allow businesses to purchase ads triggered by a user’s search for competitor trademarks (i.e., rivals’ business names).^[101] But in some contexts, this can be infringing—e.g., where the use of trademarked terms in combination with advertising text can mislead consumers about the origin of a good or service.^[102]

Thus, the harm, when it arises, would not be in a user asking an AI system to generate something “in the style of” a known creator, but when that user subsequently seeks to release a new AI-generated work and falsely claims it originated from the creator, or leaves the matter ambiguous and misleading to consumers.

Alternative remedies for creators could be found in the “right of publicity” laws in various states. A state-level right of publicity “is not merely a legal right of the ‘celebrity,’ but is a right inherent to everyone to control the commercial use of identity and persona and recover in court damages and the commercial value of an unpermitted taking.”^[103] Such rights are recognized under state common law and statutes, which vary considerably in scope across jurisdictions—frequently as part of other privacy statutes.^[104] For example, some states only protect an individual’s name, likeness, or voice, while others also cover distinctive appearances, gestures, and mannerisms.^[105] The protections afforded for right-of-publicity claims vary significantly based on the state where the unauthorized use occurs or the individual is domiciled.^[106] This creates challenges for the application of uniform nationwide protection of creators’ interests in the various aspects that such laws protect.

In recent hearings before the U.S. Senate Judiciary Subcommittee on Intellectual Property, several witnesses advocated creating a federal version of the right of publicity.^[107] The Copyright Office has also previously opined that it may be desirable for Congress to enact some form of a “right of publicity” law.^[108] If Congress chose to enact a federal “right of privacy” statute, several key issues would need to be addressed regarding the scope of protection, effect on state laws, constitutional authority, and First Amendment limitations.

Congress would have to delineate the contours of the federal right of publicity, including the aspects of identity covered and the types of uses prohibited. A broad right of privacy could protect names, images, likenesses, voices, gestures, distinctive appearances, and biographical information from any unauthorized commercial use. Or Congress could take a narrower approach focused only on particular identity attributes, like name and likeness. Congress would also need to determine whether a federal right-of-publicity statute preempts state right-of-publicity laws or sets a floor that would allow state protections to exceed the federal standards.

4.              Bargaining for the use of likenesses

A federal right of publicity could present an interesting way out of the current dispute between rightsholders and AI producers. Most of the foregoing comment attempts to pull apart different pieces of potential infringement actions, but such actions are only necessary, obviously, if a mutually beneficial agreement cannot be struck between creators and AI producers. The main issue at hand is that, given the vast amount of content necessary to train an AI system, it could be financially impractical for even the largest AI firms to license all the necessary content. Even if the comments above are correct, and fair use is not available, it could very well be the case that AI producers will not license very much content, possibly relying on public-domain material, and choosing to license only a very small selection.

Something like a “right of publicity,” or an equivalent agreement between creators and AI producers, could provide alternative licensing and monetization strategies that encourage cooperation between the parties. If creators had the opportunity to opt into the use of their likeness (or the relevant equivalent for the sort of AI system in question), the creators could generate revenue when the AI system actually uses the results of processing their content. Thus, the producers would not need to license content that contributes an unknown and possibly de minimis value to their systems, and would only need to pay for individual instances of use.

Indeed, in this respect, we are already beginning to see some experimentation with business models. The licensing of celebrity likenesses for Meta’s new AI chatbots highlights an emerging opportunity for creators to monetize their brand through contractual agreements that grant usage rights to tech companies that commercialize conversational AI.[109] As this technology matures, there will be more opportunities for collaborations between AI producers—who are eager to leverage reputable and recognizable personalities—and celebrities or influencers seeking new income streams.

As noted, much of the opportunity for creators and AI producers to reach these agreements will depend on how rights are assigned.[110] It may be the case that a “right of publicity” is not necessary to make this sort of bargaining happen, as creators could—at least theoretically—pursue litigation on a state-by-state basis. This disparate-litigation strategy could deter many creators, however, and it could also be the case that a single federal standard outlining a minimal property right in “publicity” could help to facilitate bargaining.

Conclusion

The advent of generative AI systems presents complex new public-policy challenges centered on the intersection of technology and copyright law. As the Copyright Office’s inquiry recognizes, there are open questions around the legal status of AI-training data, the attribution of AI outputs, and infringement liability, which all require thoughtful analysis.

Ultimately, maintaining incentives for human creativity, while also allowing AI systems to flourish, will require compromise and cooperation between stakeholders. Rather than an outright ban on the unauthorized use of copyrighted works for training data, a licensing market that enables access to a large corpora could emerge. Rightsholders may need to accept changes to how they typically license content. In exchange, AI producers will have to consider how they can share the benefit of their use of protected works with creators.

Copyright law retains flexibility to adapt to new technologies, as past reforms reacting to photography, sound recordings, software, and the internet all demonstrate. With careful balancing of interests, appropriate limitations, and respect for constitutional bounds, copyright can continue to promote the progress of science and the useful arts even in the age of artificial intelligence. This inquiry marks a constructive starting point, although ongoing reassessment will likely be needed as generative AI capabilities continue to advance rapidly.

[1] Artificial Intelligence and Copyright, Notice of Inquiry and Request for Comments, U.S. Copyright Office, Library of Congress (Aug. 30, 2023) [hereinafter “NOI”].

^[2] Tim Sweeney (@TimSweeneyEpic), Twitter (Jan. 15, 2023, 3:35 AM), https://twitter.com/timsweeneyepic/status/1614541807064608768?s=46&t=0MH_nl5w4PJJl46J2ZT0Dw.

^[3] Pulitzer Prize Winner and Other Authors Accuse OpenAI of Misusing Their Writing, Competition Policy International (Sep. 11, 2023), https://www.pymnts.com/cpi_posts/pulitzer-prize-winner-and-other-authors-accuse-openai-of-misusing-their-writing; Getty Images Statement, Getty Images (Jan. 17, 2023), https://newsroom.gettyimages.com/en/getty-images/getty-images-statement.

[4] See, e.g., Anton Oleinik, What Are Neural Networks Not Good At? On Artificial Creativity, 6 Big Data & Society (2019), available at https://journals.sagepub.com/doi/full/10.1177/2053951719839433#bibr75-2053951719839433.

[5] William M. Landes & Richard A. Posner, An Economic Analysis of Copyright Law, 18 J. Legal Stud. 325 (1989).

[6] Id. at 332.

[7] Id. at 326.

[8] Id.

[9] See infra, notes 102-103 and accompanying text.

^[10] See generally R.H. Coase, The Problem of Social Cost, 3 J. L. & Econ. 1, 2 (1960).

^[11] Richard Posner, Economic Analysis of Law (Aspen 5^th ed 1998) 65, 79.

^[12] Coase, supra note 9, at 27.

^[13] Id.

^[14] Id. at 27.

^[15] Id. at 42-43.

^[16] U.S. Copyright Office, Library of Congress, supra note 1, at 14.

^[17] For more detailed discussion of GANs and Stable Diffusion see Ian Spektor, From DALL E to Stable Diffusion: How Do Text-to-image Generation Models Work?, Tryo Labs Blog (Aug. 31, 2022), https://tryolabs.com/blog/2022/08/31/from-dalle-to-stable-diffusion.

^[18] Id.

^[19] Id.

^[20] Id.

^[21] Id.

^[22] Id.

^[23] Jay Alammar, The Illustrated Stable Diffusion, Blog (Oct. 4, 2022), https://jalammar.github.io/illustrated-stable-diffusion.

^[24] Indeed, there is evidence that some models may be trained in a way that they “memorize” their training set, to at least some extent. See, e.g., Kent K. Chang, Mackenzie Cramer, Sandeep Soni, & David Bamman, Speak, Memory: An Archaeology of Books Known to ChatGPT/GPT-4, arXiv Preprint (Oct. 20, 2023), https://arxiv.org/abs/2305.00118; OpenAI LP, Comment Regarding Request for Comments on Intellectual Property Protection for Artificial Intelligence Innovation, Before the USPTO, Dep’t of Com. (2019), available at https://www.uspto.gov/sites/default/files/documents/OpenAI_RFC-84-FR-58141.pdf.

^[25] OpenAI, LP, Comment Regarding Request for Comments on Intellectual Property Protection for Artificial Intelligence, id. (emphasis added).

^[26] 17 U.S.C. § 107.

^[27] See, e.g., Blake Brittain, Meta Tells Court AI Software Does Not Violate Author Copyrights, Reuters (Sep. 19, 2023), https://www.reuters.com/legal/litigation/meta-tells-court-ai-software-does-not-violate-author-copyrights-2023-09-19; Avram Piltch, Google Wants AI Scraping to be ‘Fair Use.’ Will That Fly in Court?, Tom’s Hardware (Aug. 11, 2023), https://www.tomshardware.com/news/google-ai-scraping-as-fair-use.

^[28] 17 U.S.C. § 106.

^[29] Register of Copyrights, DMCA Section 104 Report (U.S. Copyright Office, Aug. 2001), at 108-22, available at https://www.copyright.gov/reports/studies/dmca/sec-104-report-vol-1.pdf.

^[30] Id. at 122-23.

^[31] Id. at 112 (emphasis added).

[32] Id. at 129–30.

[33] 17 U.S.C. § 107.

[34] Id.; see also Campbell v. Acuff-Rose Music Inc., 510 U.S. 569 (1994).

[35] Critically, a fair use analysis is a multi-factor test, and even within the first factor, it’s not a mandatory requirement that a use be “transformative.” It is entirely possible that a court balancing all of the factors could indeed find that training AI systems is fair use, even if it does not hold that such uses are “transformative.”

[36] Campbell, supra note 22, at 591.

[37] Authors Guild v. Google, Inc., 804 F.3d 202, 214 (2d Cir. 2015).

[38] OpenAI submission, supra note 13, at 5.

[39] Id. at 915.

[40] Id.

[41] Id.

^[42] Id. at 933-34.

^[43] Id. at 923. (emphasis added)

^[44] Id.

[45] Id. at 924.

[46] Kelly v. Arriba Soft Corp., 336 F.3d 811 (9th Cir. 2002).

[47] Id.

[48] Id. at 818.

[49] Id.

[50] Id. at 819 (“Arriba’s use of the images serves a different function than Kelly’s use—improving access to information on the internet versus artistic expression.”).

[51] The “public benefit” aspect of copyright law is reflected in the fair-use provision, 17 U.S.C. § 107. In Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994), the Supreme Court highlighted the “social benefit” that a use may provide, depending on the first of the statute’s four fair-use factors, the “the purpose and character of the use.”

[52] Supra note 46, at 820.

[53] Perfect 10 Inc. v. Amazon.com Inc., 487 F.3d 701 (9^th Cir., 2007)

[54] Id. at 721 (“Although an image may have been created originally to serve an entertainment, aesthetic, or informative function, a search engine transforms the image into a pointer directing a user to a source of information.”).

[55] Id. at 721.

[56] Id. at 723 (emphasis added).

[57] Id. (emphasis added).

[58] Id.

[59] Supra note 37, at 218.

[60] Id. at 215-16.

[61] Id. at 214. See also id. (“The more the appropriator is using the copied material for new, transformative purposes, the more it serves copyright’s goal of enriching public knowledge and the less likely it is that the appropriation will serve as a substitute for the original or its plausible derivatives, shrinking the protected market opportunities of the copyrighted work.”).

[62] Id. at 218.

[63] Perfect 10, 487 F.3d at 721-22 (citing Kelly, 336 F.3d at 818-19). See also Campbell, 510 U.S. at 579 (“The central purpose of this investigation is to see, in Justice Story’s words, whether the new work merely ‘supersede[s] the objects’ of the original creation, or instead adds something new, with a further purpose or different character….”) (citations omitted).

[64] See supra, notes 9-14 and accompanying text.

[65] See, e.g., the development of the compulsory “mechanical royalty,” now embodied in 17 U.S.C. § 115, that was adopted in the early 20^th century as a way to make it possible for the manufacturers of player pianos to distribute sheet music playable by their instruments.

[66] See supra notes 9-14 and accompanying text.

^[67] U.S. Copyright Office, Library of Congress, supra note 1, at 15.

[68] See infra, notes at 102-103 and accompanying text.

^[69] U.S. Copyright Office, Library of Congress, supra note 1, at 15.

^[70] See, e.g., Copyright Free Music, Premium Beat By Shutterstock, https://www.premiumbeat.com/royalty-free/licensed-music; Royalty-free stock footage at your fingertips, Adobe Stock, https://stock.adobe.com/video.

^[71] U.S. Copyright Office, Library of Congress, supra note 1, at 19.

^[72] Id.

^[73] U.S. Const. art. I, § 8, cl. 8.

[74] See Ajay Agrawal, Joshua S. Gans, & Avi Goldfarb, Exploring the Impact of Artificial Intelligence: Prediction Versus Judgment, 47 Info. Econ. & Pol’y 1, 1 (2019) (“We term this process of understanding payoffs, ‘judgment’. At the moment, it is uniquely human as no machine can form those payoffs.”).

^[75] Joshua Gans, Can AI works get copyright protection? (Redux), Joshua Gans’ Newsletter (Sept. 7, 2023), https://joshuagans.substack.com/p/can-ai-works-get-copyright-protection.

^[76] See Nicholas Carlini, et al., Extracting Training Data from Diffusion Models, Cornell Univ. (Jan. 30, 2023), available at https://arxiv.org/abs/2301.13188.

^[77] See Chang, Cramer, Soni, & Bamman, supra note 24; see also Matthew Sag, Copyright Safety for Generative AI, Working Paper (May 4, 2023), available at https://ssrn.com/abstract=4438593.; Andrés Guadamuz, A Scanner Darkly: Copyright Liability and Exceptions in Artificial Intelligence Inputs and Outputs, 25-27 (Mar. 1, 2023), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4371204.

^[78] Laureyssens v. Idea Grp. Inc., 964 F.2d 131, 140 (2d Cir. 1992), as amended (June 24, 1992).

^[79] Id. at 139.

^[80] Harney v. Sony Pictures Television Inc., 704 F.3d 173, 178 (1st Cir. 2013). This assumes, for argument’s sake, that a given model is not “memorizing,” as noted above.

^[81] Id. at 178-79.

^[82] Id.

[83] U.S. Copyright Office, Library of Congress, supra note 1, at 25.

^[84] Notably, the state of mind of the user would be irrelevant from the point of view of whether an infringement occurs. All that is required is that a plaintiff owns a valid copyright, and that the defendant infringed it. 17 U.S.C. 106. There are cases where the state of mind of the defendant will matter, however. For one, willful or recklessly indifferent infringement by a plaintiff will open the door for higher statutory damages. See, e.g., Island Software & Computer Serv., Inc. v. Microsoft Corp., 413 F.3d 257, 263 (2d Cir. 2005). For another, a case of criminal copyright infringement will require that a defendant have acted “willfully.” 17 U.S.C. § 506(a)(1) (2023), 18 U.S.C. § 2319 (2023).

^[85] Id. at 550.

[86] Legally speaking, it would be incoherent to suggest that there can be secondary liability without primary liability. The way that AI systems work, however, could prompt Congress to modify the law in order to account for the identified situation.

^[87] See, e.g., Metro-Goldwyn-Mayer Studios Inc. v. Grokster Ltd., 380 F.3d 1154, 1160 (9th Cir. 2004), vacated and remanded, 545 U.S. 913, 125 S. Ct. 2764, 162 L. Ed. 2d 781 (2005).

^[88] See BMG Rts. Mgmt. (US) LLC v. Cox Commc’ns Inc., 881 F.3d 293, 306 (4th Cir. 2018); Sony Corp. of Am. v. Universal City Studios Inc., 464 U.S. 417, 442 (1984).

^[89] A&M Recs. Inc. v. Napster Inc., 239 F.3d 1004, 1022 (9th Cir. 2001), as amended (Apr. 3, 2001), aff’d sub nom. A&M Recs. Inc. v. Napster Inc., 284 F.3d 1091 (9th Cir. 2002), and aff’d sub nom. A&M Recs. Inc. v. Napster Inc., 284 F.3d 1091 (9th Cir. 2002).

^[90] See, e.g., Content Filtering, Microsoft Ignite, available at https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/content-filter (last visited Oct. 27, 2023).

^[91] Note that, if an AI producer can demonstrate that they used no protected works in the training phase, there may in fact be no liability for infringement at all. If a protected work is never made available to the AI system, even an output very similar to that protected work might not be “substantially similar” in a legal sense.

^[92] Copyright Review Board, Second Request for Reconsideration for Refusal to Register Théâtre D’opéra Spatial (SR # 1-11743923581; Correspondence ID: 1-5T5320R), U.S. Copyright Office (Sep. 5, 2023), available at https://fingfx.thomsonreuters.com/gfx/legaldocs/byprrqkqxpe/AI%20COPYRIGHT%20REGISTRATION%20decision.pdf.

^[93] Am. Broad. Companies Inc. v. Aereo Inc., 573 U.S. 431, 453 (2014). (Thomas J, dissenting).

^[94] Id. at 451.

[95] U.S. Copyright Office, Library of Congress, supra note 1, at 21.

^[96] See 5 U.S.C. § 1051 et seq. at § 1127.

^[97] See, e.g., ETW Corp. v. Jireh Pub. Inc., 332 F.3d 915, 923 (6th Cir. 2003).

^[98] Dastar Corp. v. Twentieth Century Fox Film Corp., 539 U.S. 23, 34 (2003).

^[99] Id. at 37.

^[100] Id. at 32.

^[101] See, e.g., Multi Time Mach. Inc. v. Amazon.com Inc., 804 F.3d 930, 938 (9^th Cir. 2015); EarthCam Inc. v. OxBlue Corp., 49 F. Supp. 3d 1210, 1241 (N.D. Ga. 2014); Coll. Network Inc. v. Moore Educ. Publishers Inc., 378 F. App’x 403, 414 (5th Cir. 2010).

^[102] Digby Adler Grp. LLC v. Image Rent a Car Inc., 79 F. Supp. 3d 1095, 1102 (N.D. Cal. 2015).

^[103] J. Thomas McCarthy, The Rights of Publicity and Privacy § 1:3. Introduction—Definition and History of the Right of Publicity—Simple Definition of the Right of Publicity, 1 Rights of Publicity and Privacy § 1:3 (2d ed).

^[104] See id. at § 6:3.

^[105] Compare Ind. Code § 32-36-1-7 (covering name, voice, signature, photograph, image, likeness, distinctive appearance, gesture, or mannerism), with Ky. Rev. Stat. Ann. § 391.170 (limited to name and likeness for “public figures”).

^[106] See Restatement (Third) of Unfair Competition § 46 (1995).

^[107] See, e.g., Jeff Harleston, Artificial Intelligence and Intellectual Property – Part II: Copyright, U.S. Senate Comm. on the Judiciary Subcomm. on Intellectual Property (Jul.12, 2023), available at https://www.judiciary.senate.gov/imo/media/doc/2023-07-12_pm_-_testimony_-_harleston1.pdf; Karla Ortiz, “AI and Copyright”, U.S. Senate Comm. on the Judiciary Subcomm. on Intellectual Property (Jul. 7, 2023), available at https://www.judiciary.senate.gov/imo/media/doc/2023-07-12_pm_-_testimony_-_ortiz.pdf; Matthew Sag, “Artificial Intelligence and Intellectual Property – Part II: Copyright and Artificial Intelligence”, U.S. Senate Comm. on the Judiciary Subcomm. on Intellectual Property (Jul. 12, 2023), available at https://www.judiciary.senate.gov/imo/media/doc/2023-07-12_pm_-_testimony_-_sag.pdf.

^[108] Authors, Attribution, and Integrity: Examining Moral Rights in the United States, U.S. Copyright Office (Apr. 2019) at 117-119, https://www.copyright.gov/policy/moralrights/full-report.pdf.

[109] Benj Edwards, Meta Launches Consumer AI Chatbots with Celebrity Avatars in its Social Apps, ArsTechnica (Sep. 28, 2023), https://arstechnica.com/information-technology/2023/09/meta-launches-consumer-ai-chatbots-with-celebrity-avatars-in-its-social-apps; Max Chafkin, Meta’s New AI Buddies Aren’t Great Conversationalists, Bloomberg (Oct. 17, 2023), https://www.bloomberg.com/news/newsletters/2023-10-17/meta-s-celebrity-ai-chatbots-on-facebook-instagram-are-surreal.

[110] See supra, notes 8-14 and accompanying text.