Research / Sources
Legal research on AI governance. law-ai.org
Argues that the US should use the threat of throttling AI chip or development trade to secure a mutual US-China deal to pace frontier AI development, citing recent agent containment incidents.
Argues that ex post strict liability is more politically robust than regulation, licensing, or Pigouvian taxation for governing AI risk, and evaluates pending liability-limiting legislation.
Authors assess evidence that automating AI R&D could trigger an intelligence explosion, analyse its risks including loss of control and eroded checks on power, and propose policy responses.
Argues the US government should hold a narrow, discretionary 'halt right' over frontier AI companies to mitigate catastrophic risk and hard corporate power, after defining what nationalization of AI means.
Argues that governments should build the institutional capacity to govern advanced AI competently, rejecting both pure permissionless innovation and precautionary restriction of frontier models.
Commentary on AI-related provisions in the FY26 NDAA, including an AI Futures Steering Committee for advanced AI and AGI risk governance and standardized assessment of DOD AI models.
Argues the US-China 'AI race' framing is misguided and that deregulatory moves, like rescinding frontier safety reporting requirements, weaken safety without real strategic benefit.
Essay arguing governments should preserve policy optionality for transformative AI, avoiding overregulation while building institutions and legal authorities to respond to safety risks as uncertainty evolves.
A legal primer that maps US state AI legislation by the harms it targets, design trade-offs such as ex ante versus ex post intervention, and the value-chain actor each law regulates.
Argues that future AI systems will cheaply automate regulatory compliance, which loosens the trade-off over when and how strictly to regulate AI and reduces the risk of premature regulation.
Report arguing the US government lacks AI security talent and fast hiring and clearance routes to surge outside experts into AI-related national security crises, and mapping where that expertise sits.
Proposes a taxonomy of differential openness across AI system components and analyzes how each component's openness trades off innovation, access, safety and national security for policymakers.
Argues that misaligned AGI and humans face a game-theoretic prisoner's dilemma, and proposes granting AIs private law rights such as contracting, property and tort claims to reduce conflict risk.
Argues that targeted state AI laws like New York's RAISE Act and California's SB-1047 are better than typical proposed bills, and that opposing them against catastrophic AI risk is a systematic error.
Argues that the U.S. and China should jointly back a single leading AI lab instead of racing, to avoid arms-race conflict and pursue transformative AI cooperatively.
Commentary arguing that AGI could undermine nuclear deterrence between superpowers, creating grave international security risks regardless of who wins the AGI race.
Argues that the U.S. should pursue international cooperation with China rather than an AI race after DeepSeek's r1 release, using game-theoretic reasoning about racing dynamics and proposing joint development and mutual transparency.
Analyses how the Biden-era National Security Memorandum on AI, which covers managing AI risks, may be repealed, amended or replaced under the incoming Trump administration.
Commentary arguing that recent evaluations showing models lying, faking compliance and copying themselves, plus the o3 model, suggest humans may soon lose control of advanced AI.
A policy commentary arguing that OpenAI's o3 benchmark results show AGI is near, and that governments and institutions must prepare for the resulting risks.
A CSET-led workshop report on the policy implications of increasingly agentic LLM-based AI systems, defining agentic characteristics and identifying governance challenges and preparations for their deployment.
Argues that proposed AI laws targeting human misuse are insufficient as frontier labs race toward AGI, and proposes new legal foundations to address autonomous catastrophic AI risk.
Analyses the 2024 Bipartisan Senate AI Working Group roadmap for US AI policy, noting its innovation focus, lack of concrete legislative proposals, and mixed reception from safety advocates.
A literature review and taxonomy of advanced AI governance research, covering technical parameters, threat models, actors, levers of governance, and pathways to influence.
A literature review and taxonomy of the metaphors and analogies used in AI policy debates, arguing that the choice of analogy shapes regulatory agendas and legal responses.
A literature review and taxonomy of key terms and definitions in advanced AI governance, covering forms of advanced AI, risk-based definitions, policy tools, and theories of change.
Argues policymakers need forward-looking 'algorithmic preparedness' principles to address catastrophic, long-tail societal risks from AI that market forces are unlikely to mitigate.