Raymond on Our AI, Ourselves: Illuminating the Human Fears Animating Early Regulatory Responses to the Use of Generative AI in the Practice of Law

Margaret Raymond (U Wisconsin Law) has posted “Our AI, Ourselves: Illuminating the Human Fears Animating Early Regulatory Responses to the Use of Generative AI in the Practice of Law” (15 St. Mary’s Journal on Legal Malpractice & Ethics 221 (2025)) on SSRN. Here is the abstract:

Generative artificial intelligence is changing the way lawyers work, and with those changes have come questions and concerns about how it should be regulated. Those questions and concerns, particularly on the individual level, are driven by fears about the implications of the use of generative AI. This Article identifies and explores the fears that drive these regulatory responses: fear of exposing judicial fallibility, anxiety over AI replacing human lawyers, and concerns about missing out on AI’s potential benefits. Ultimately, effective regulation of the use of generative AI in legal practice needs to be attentive to the fears and hopes surrounding generative AI in law. Only by understanding the very human anxieties regarding generative AI can the profession craft effective regulatory models that address the integration of AI in legal practice.

Mei et al. on The Illusory Normativity of Rights-Based AI Regulation

Yiyang Mei (Emory U) and Matthew Sag (Emory U Law) have posted “The Illusory Normativity of Rights-Based AI Regulation” on SSRN. Here is the abstract:

Whether and how to regulate AI is now a central question of governance. Across academic, policy, and international legal circles, the European Union is widely treated as the normative leader in this space. Its regulatory framework, anchored in the General Data Protection Regulation, the Digital Services and Markets Acts, and the AI Act, is often portrayed as a principled model grounded in fundamental rights. This Article challenges that assumption. We argue that the rights-based narrative surrounding EU AI regulation mischaracterizes the logic of its institutional design. While rights language pervades EU legal instruments, its function is managerial, not foundational. These rights operate as tools of administrative ordering, used to mitigate technological disruption, manage geopolitical risk, and preserve systemic balance, rather than as expressions of moral autonomy or democratic consent. Drawing on comparative institutional analysis, we situate EU AI governance within a longer tradition of legal ordering shaped by the need to coordinate power across fragmented jurisdictions. We contrast this approach with the American model, which reflects a different regulatory logic rooted in decentralized authority, sectoral pluralism, and a constitutional preference for innovation and individual autonomy. Through case studies in five key domains—data privacy, cybersecurity, healthcare, labor, and disinformation—we show that EU regulation is not meaningfully rights-driven, as is often claimed. It is instead structured around the containment of institutional risk. Our aim is not to endorse the American model but to reject the presumption that the EU approach reflects a normative ideal that other nations should uncritically adopt. The EU model is best understood as a historically contingent response to its own political conditions, not a template for others to blindly follow.

Lyons on The Litigation Solution: Why Courts, Not Code Mandates, Should Address AI Discrimination

Daniel Lyons (Boston College Law) has posted “The Litigation Solution: Why Courts, Not Code Mandates, Should Address AI Discrimination” on SSRN. Here is the abstract:

As artificial intelligence systems increasingly influence decisionmaking in high-stakes sectors, policymakers have focused on regulating model design to combat algorithmic bias. Drawing on examples from the European Union’s AI Act and recent state legislation, this Article critiques the emerging “fairness by design” paradigm. It argues that design mandates rest on a flawed premise: that bias can be objectively defined and mitigated ex ante without compromising competing values such as accuracy, privacy, or innovation. In reality, efforts to engineer fairness through prescriptive regulation risk distorting markets, entrenching incumbents, and stifling technological advancement. Moreover, the opaque, evolving nature of AI systems—especially generative models—makes it difficult to anticipate or eliminate future biases through design alone, often creating tradeoffs that regulators are ill-equipped to manage.

Rather than regulating AI inputs, the Article advocates for a litigation-first approach that focuses on AI outputs and leverages existing antidiscrimination law to address harms as they arise. By applying traditional disparate treatment and disparate impact frameworks to AI-assisted decisions, courts can assess when biased outcomes rise to the level of unlawful discrimination—without prematurely constraining innovation or imposing rigid mandates. This model mirrors America’s historical preference for permissive innovation, allowing technology to evolve while holding bad actors accountable under general principles of law. The result is a more flexible, targeted regulatory regime that fosters AI development while safeguarding civil rights.

Klonowska et al. on Rhetoric and Regulation: The (Limits of) Human/AI Comparison in Legal Debates on Military AI

Klaudia Klonowska (T.M.C. Asser Institute) and Taylor Kate Woodcock (T.M.C. Asser Institute) have posted “Rhetoric and Regulation: The (Limits of) Human/AI Comparison in Legal Debates on Military AI” (Forthcoming in Boutin B., Woodcock T. K. & Soltanzadeh S. (eds.), Decision at the Edge: Interdisciplinary Dilemmas in Military Artificial Intelligence, Asser Press (2025)) on SSRN. Here is the abstract:

The promise of artificial intelligence (AI) is ubiquitous and compelling, yet can it truly deliver ‘better’ speed, accuracy, and decision making in the conduct of war? As AI becomes increasingly embedded in targeting processes, legal and ethical debates often compare who performs better, humans or machines? In this Chapter, we unpack and critique the prevalence of comparisons between humans and AI systems, including in analyses of the fulfilment of legal obligations under International Humanitarian Law (IHL). We challenge this binary framing by highlighting misleading assumptions that neglect how the use of AI results in complex human-machine interactions that transform targeting practices. We unpack what is meant by ‘better performance’, demonstrating how prevailing metrics for speed and accuracy can create misleading expectations around the use of AI given the realities of warfare. We conclude that holistic but granular attention must be paid to the landscape of human-machine interactions to understand how the use of AI impacts compliance with IHL targeting obligations.

Feher et al. on Is AI Trained on Public Money? Evidence from US Data Centers

Adam Feher (U Lausanne) et al. have posted “Is AI Trained on Public Money? Evidence from US Data Centers” on SSRN. Here is the abstract:

Rapid data center growth has raised concerns about rising energy demand and its effects. Leveraging a novel dataset of U.S. data center energy loads, utility prices, and establishment-level outcomes, we quantify local spillover effects on electricity prices, firm performance, and emissions. Using an IV continuous DiD exploiting exogenous variation in data center location attractiveness, we find no local spillovers over 2010–2024. A regional model calibrated to the empirical null suggests that shocks larger than those observed through 2024 could still result in noticeable increases in household utility bills if not offset by regulation or external supply.

Fan et al. on Novel Corporate Governance Structures

Jennifer S. Fan (Loyola Law Los Angeles) and Xuan-thao Nguyen (U Washington Law) have posted “Novel Corporate Governance Structures” (Harvard Journal of Law & Technology, Volume 38, Number 4 Spring 2025) on SSRN. Here is the abstract:

Artificial Intelligence (“AI”) startups have taken center stage, rapidly disrupting conventional industries at an unprecedented pace with their groundbreaking innovations. Hailed by many as the most significant technological advancement of our era, AI’s profound societal impact has garnered heightened public and governmental scrutiny. The spotlight has recently fallen on OpenAI, the creator of ChatGPT, which weathered a tumultuous period marked by the ouster and subsequent rehiring of CEO Sam Altman, a board reconfiguration, and Altman’s later return to the board. Concerns over AI safety were offered as the rationale for the tandem corporate governance structure of nonprofit and for-profit at OpenAI which led to board friction, a management coup, and superalignment defection. Similarly, concerns over AI safety also underscore the creation of the corporate structures at Anthropic and xAI.

This Article explores the innovative corporate governance models that have emerged from leading AI startups like OpenAI, Anthropic, and xAI, assessing their long-term viability as these companies race against one another in building AI foundation models. Ultimately, it proposes a path forward for improved governance in AI startups by advocating for an amendment to corporate law requiring a board-level AI Safety Committee at AI startups.

Barnett on The Free Content Illusion

Jonathan Barnett (USC Gould Law) has posted “The Free Content Illusion” (Journal of Intellectual Property Law (2026)) on SSRN. Here is the abstract:

Peer-to-peer file sharing in the early 2000s destabilized traditional content markets and associated business models that rely on preserving control over the use of creative assets. Academics and other commentators widely argued that robust forms of intellectual property rights had been rendered largely obsolete in a digital environment of low production and distribution costs. Reflecting this view, courts expanded the fair use doctrine and generously applied safe harbors under the Digital Millenium Copyright Act, which largely immunized platforms against liability for user infringement and consistently favored content aggregators over originators. The subsequent evolution of digital markets nonetheless shows that exclusivity protections remain critical to sustaining an independently viable content economy that does not rely on philanthropic or governmental patronage. Streaming services in audio, video, and literary media restored revenue flows to content originators through contractual and technological complements to copyright protection, while content segments (notably, the news industry) that failed to deploy such mechanisms struggled economically. Contrary to prevailing views, meaningful property rights and other exclusivity protections remain essential for sustaining the production, financing, and development of creative assets in digital environments and, together with technological and contractual complements, are likely to retain this role in supporting a robust flow of original content for the artificial intelligence ecosystem.