Executive Summary
As artificial intelligence diffuses throughout society, policymakers are faced with the challenge of how best to govern the technology amid uncertainty over the future of AI development. To meet this challenge, many stakeholders have put forth various proposals aimed at shaping AI governance approaches. This report outlines an analytic approach to help policymakers make sense of such proposals and take steps to govern AI systems while preserving future decision-making flexibility. Our approach involves analyzing common assumptions across various proposals (as these assumptions are foundational elements for the success of multiple proposals), as well as unique assumptions across individual proposals, by answering three questions:
- What risks are important to mitigate and who should have primary oversight of frontier AI?
- Who is delegated tasks and able to play a role?
- Would the proposed mechanisms or tools actually achieve the proposal’s objectives?
We apply this analytic approach to five U.S.-centric AI governance proposals that originate from industry, academia, civil society, and the federal and state governments. These proposals are generally aimed at governing frontier AI systems, which possess cutting-edge capabilities and therefore pose some of the most challenging questions for AI governance. Our analysis reveals that most proposals view AI-enabling talent and AI processes and frameworks as important enablers of AI governance. However, proposals lack consensus regarding the techniques that are most effective at mitigating AI risks and harms.
Our analysis also bears lessons that are broadly applicable to policymakers seeking to analyze any proposal. Our case studies demonstrate that 1) policymakers should leverage proposals’ assumptions to more precisely understand disagreements and shared views among stakeholders and 2) policymakers can take action in an uncertain and rapidly changing environment by addressing common assumptions across proposals. By adopting our analytic approach, U.S. policymakers can move away from rhetorical debates about AI governance and better prepare the United States for a range of possible AI futures.