Brooklyn Law Review Proposes Dual-Board Corporate Governance Model to Make AI Safety Obligations Enforceable at Board Level

A peer-reviewed article published in the Brooklyn Law Review, titled "Governing Artificial Intelligence: A Dual-Board Solution," proposes a structural reform to U.S. corporate governance that would require companies deploying consequential AI systems to establish a second, specialized board with explicit responsibility for AI safety oversight. The proposal applies within the U.S. jurisdiction and envisions that this AI safety board would operate alongside the traditional board of directors, holding distinct fiduciary duties focused on identifying, monitoring, and mitigating AI-related harms. The article frames AI safety not as a discretionary risk management concern but as an obligation that can and should be embedded in the evolving body of corporate governance law, potentially enforceable through shareholder litigation and regulatory scrutiny. The authors argue that this structural separation is necessary to give AI safety concerns institutional standing comparable to financial and audit oversight, which have long benefited from dedicated committee structures and legal accountability.

The article emerges in the context of growing pressure on corporate boards to demonstrate meaningful AI oversight, a trend reflected in guidance from the SEC and shareholder activism around AI risk disclosures. Traditional single-board models, even those with designated AI or technology committees, lack the structural independence and specialized accountability that the authors argue AI safety demands. This proposal draws on parallels with the audit committee model that emerged from Sarbanes-Oxley, where Congress mandated a distinct substructure within boards to address a specific category of systemic risk. The article also engages with broader debates in AI governance, including questions about whether voluntary frameworks such as the NIST AI Risk Management Framework or the ISO 42001 management system standard provide sufficient enforcement mechanisms at the corporate level, and whether legal liability structures need to be redesigned to create genuine board-level incentives for AI safety investment.

For enterprise compliance teams, this article is relevant not only as scholarly analysis but as an early signal of the legal theories that plaintiffs' attorneys, regulators, and institutional shareholders may begin to adopt when evaluating board-level accountability for AI incidents. General counsel and chief compliance officers should assess whether their current board committee structures would withstand scrutiny under a dual-board standard, particularly in sectors with high-stakes AI deployments such as financial services, healthcare, and infrastructure. Risk and governance teams should consider whether their existing AI governance documentation, including risk registers, safety testing records, and incident response protocols, would demonstrate the kind of board-level engagement the article envisions as legally adequate. Companies subject to SEC reporting requirements or state-level AI governance statutes such as the Colorado AI Act should also evaluate whether evolving scholarly frameworks like this one are influencing regulatory staff thinking on what constitutes adequate AI oversight at the board level.