The Nuclear Analogy in AI Governance Research

TL;DR

This paper reviews how the nuclear analogy influences AI governance research, highlighting its role in shaping policy frameworks, strategic analysis, and cautionary lessons despite domain differences.

Contribution

It systematically analyzes 43 scholarly works applying nuclear precedents to AI governance, clarifying their functions and implications for policy development.

Findings

Nuclear analogies help analyze strategic dynamics in AI governance.

Historical lessons from nuclear policy inform AI safety and regulation.

Analogies expand policy options and legitimize radical proposals.

Abstract

The analogy between Artificial Intelligence (AI) and nuclear weapons is prominent in academic and policy discourse on AI governance. This chapter reviews 43 scholarly works which explicitly draw on the nuclear domain to derive lessons for AI governance. We identify four problem areas where researchers apply nuclear precedents: (1) early development and governance of transformative technologies; (2) international security risks and strategy; (3) international institutions and agreements; and (4) domestic safety regulation. While nuclear-inspired AI proposals are often criticised due to differences across domains, this review clarifies how historical analogies can inform policy development even when technological domains differ substantially. Valuable functions include providing conceptual frameworks for analyzing strategic dynamics, offering cautionary lessons about unsuccessful governance approaches, and expanding policy imagination by legitimizing radical proposals. Given that policymakers already invoke the nuclear analogy, continued critical engagement with these historical precedents remains essential for shaping effective global AI governance.