Cognitive Revolution Podcast: A Market for AI Regulation?

Dr. Gillian Hadfield from Johns Hopkins University and Andrew Freedman from Fathom discuss their innovative proposal to govern AI through private regulatory markets, which has been introduced as California's SB 813. Their system would separate democratic goal-setting from technical rule-making by having government bodies articulate safety outcomes while competitive private certifiers develop and enforce detailed standards, with companies receiving liability protection for compliance. The conversation explores how this market-based approach could create a "race to the top" in AI safety standards while remaining agile enough to keep pace with rapid technological development. Key challenges discussed include preventing a race to the bottom among certifiers, liability law interactions, and identifying qualified organizations to serve as effective private regulators.