Chapter 13 Further Reading: Governing AI — Policy, Regulation, and Global Approaches

Essential Reading (Start Here)

Anu Bradford, The Brussels Effect: How the European Union Rules the World (2020) Bradford's concept of the "Brussels Effect" — the EU's ability to set global regulatory standards through the sheer size of its market — is essential for understanding why the EU AI Act matters beyond Europe. The book is about regulatory influence more broadly, but its framework directly applies to AI governance. If you want to understand why companies worldwide are adjusting their practices to comply with EU AI rules even if they are not based in Europe, this is the place to start.

Gary Marchetti and Wendell Wallach, "Coordinating Technology Governance" (2022, Issues in Science and Technology) A thoughtful overview of the challenges of governing emerging technologies, with specific attention to AI. The authors argue for adaptive governance mechanisms that can keep pace with technological change — directly relevant to the pacing problem discussed in this chapter.

The EU AI Act (full text, with explanatory summaries) The actual regulation is available online and, while dense, is surprisingly readable for legal text. For a more accessible overview, the Future of Life Institute maintains a comprehensive summary and explainer at artificialintelligenceact.eu, with the full text organized by chapter and searchable by topic.

NIST AI Risk Management Framework (AI RMF 1.0, 2023) The U.S. National Institute of Standards and Technology's voluntary framework for managing AI risks. At 42 pages plus supplementary materials, it is the most concrete articulation of the U.S. approach to AI governance. It is voluntary, but it is increasingly referenced by federal agencies and state legislatures. Available at: nist.gov/artificial-intelligence/ai-risk-management-framework

Deeper Exploration

Frank Pasquale, The Black Box Society: The Secret Algorithms That Control Money and Information (2015) Published before the current AI governance wave, Pasquale's book anticipated many of the transparency and accountability challenges we now face. His argument that algorithmic decision-making systems operate as opaque "black boxes" with enormous power over people's lives remains one of the most influential framings in the field.

Anu Bradford, Digital Empires: The Global Battle to Regulate Technology (2023) Bradford's follow-up to The Brussels Effect directly examines the divergent regulatory approaches of the U.S., EU, and China to digital technology, including AI. Her typology of "market-driven" (U.S.), "rights-driven" (EU), and "state-driven" (China) models maps closely onto the comparative analysis in this chapter.

Meredith Whittaker, "The Steep Cost of Capture" (2021, Interactions) An essay by the co-founder of the AI Now Institute examining how the concentration of AI development in a few large companies affects governance. Whittaker argues that the AI industry's structure — where a handful of companies control the compute, data, and talent necessary for cutting-edge AI — creates unique governance challenges that cannot be addressed by regulation alone.

Jack Clark and Gillian Hadfield, "Regulatory Markets for AI Safety" (2024, working paper) A provocative proposal for "regulatory markets" — a system where multiple private regulatory bodies compete to provide oversight of AI systems, with government setting the standards they must meet. Whether or not you agree with the proposal, it is a creative attempt to address the pacing problem and the knowledge gap simultaneously.

Reports and Policy Documents

OECD AI Policy Observatory (oecd.ai) The Organisation for Economic Co-operation and Development maintains the most comprehensive database of AI policy initiatives worldwide. Their interactive dashboard tracks AI policies by country, policy area, and governance approach. An invaluable resource for comparative AI governance analysis.

AI Now Institute Annual Reports (ainowinstitute.org) The AI Now Institute at New York University publishes annual reports on the social implications of AI, with substantial sections on governance. Their reports are accessible, well-sourced, and consistently identify governance gaps and emerging challenges before they become mainstream concerns.

Stanford HAI, "Artificial Intelligence Index Report" (annual) Stanford's Human-Centered AI Institute publishes the most comprehensive annual survey of AI development, including a dedicated chapter on AI governance and policy. The report tracks legislation, regulation, and governance initiatives worldwide, providing quantitative data on the pace and scope of AI governance activity.

European Commission, "Laying Down Harmonised Rules on Artificial Intelligence" (2024) The official text of the EU AI Act, along with supporting documents, impact assessments, and implementation guidelines. Dense but authoritative. For a more accessible entry point, use the Future of Life Institute's annotated guide (artificialintelligenceact.eu) or the European Commission's official FAQ documents.

White House Executive Order on AI (October 2023) The full text of President Biden's executive order on AI, which directed federal agencies to develop safety standards, required AI developers to share safety test results, and established principles for government AI use. Available at whitehouse.gov. An important document for understanding the U.S. approach, though its durability depends on subsequent presidential administrations.

Perspectives and Debates

Azeem Azhar, The Exponential Age (2021) Azhar's thesis that technology is growing exponentially while institutions change incrementally provides an accessible framework for understanding the pacing problem. While not exclusively about AI governance, the book's core argument — that the "exponential gap" between technology and governance is widening — is directly relevant.

Cass Sunstein, "The Precautionary Principle as a Basis for Decision Making" (2003, The Economists' Voice) A concise examination of the precautionary principle — the idea that technologies should be proven safe before deployment — by one of America's most influential legal scholars. Sunstein raises important objections that complicate simplistic applications of precaution to AI governance.

Jobin, Ienca, and Vayena, "The Global Landscape of AI Ethics Guidelines" (2019, Nature Machine Intelligence) A systematic analysis of 84 AI ethics guidelines from around the world, identifying common themes and significant divergences. The study found broad consensus on principles like transparency and fairness but much less agreement on how those principles should be implemented. A useful reference for the gap between aspirational principles and practical governance.

Daron Acemoglu, "The Simple Macroeconomics of AI" (2024, NBER Working Paper) A leading economist's analysis of AI's economic implications, including the policy responses needed to ensure broad-based benefit. While primarily economic rather than regulatory, Acemoglu's argument that AI's benefits depend heavily on the institutional and policy environment provides important context for governance debates.

Multimedia Resources

"The AI Dilemma" — Center for Humane Technology (2023, presentation) A presentation by Tristan Harris and Aza Raskin on the governance challenges posed by rapidly advancing AI. Available on YouTube. While some claims are contested, the presentation effectively communicates the urgency of AI governance to a general audience.

Lawfare Blog, AI section (lawfaremedia.org) The Lawfare blog, run by the Brookings Institution, covers the legal and policy dimensions of AI in depth. Their AI coverage includes analysis of legislation, regulatory actions, court decisions, and international governance developments. Consistently high quality and accessible to non-lawyers.

CSET (Center for Security and Emerging Technology) reports (cset.georgetown.edu) Georgetown University's CSET publishes policy-focused research on AI governance, with particular strength in the geopolitical dimensions — the U.S.-China AI competition, international AI governance, and the intersection of AI with national security. Their reports are well-researched and written for a policy audience.

For Your AI Audit Report

If you are researching the governance landscape for your chosen AI system, these resources may be particularly useful:

• Use the OECD AI Policy Observatory to identify which countries' regulations apply to your system and what policy initiatives are relevant.
• Reference the EU AI Act text (via the Future of Life Institute explainer) to classify your system under the risk-based framework.
• Consult the NIST AI RMF for a structured approach to identifying and managing risks associated with your system.
• Check the Stanford HAI AI Index for the latest data on governance trends in your system's sector.
• Review your system's developer or operator's published AI principles (if any), and use the Jobin et al. study as a framework for evaluating how those principles compare to global norms.