Building Technology to Drive AI Governance: Measurement and Cost Reduction

The article argues that technology plays a crucial role in driving AI governance by shifting the underlying dynamics of AI development. It highlights two key technological mechanisms: measurement and cost reduction. Measurement creates visibility, enables accountability, and makes regulation feasible, while driving down costs makes good behavior economically practical and can dissolve apparent trade-offs. The author draws parallels between AI and other domains like climate change and food safety, where these mechanisms have proven effective.

In the context of climate change, improved measurement through global monitoring of temperature and CO₂ levels has informed strategy and galvanized action. Satellite imagery of site-level emissions has shifted incentives by making leaks measurable and holding companies accountable. Similarly, in food safety, measurement and cost reduction have driven better outcomes. The author suggests that building technologies for AI governance, such as those that improve measurement and reduce costs, is currently the most leveraged thing technically skilled people can do. This approach is considered more effective than direct technical work that ignores governance or policy work that is untethered from technical solutions. The article emphasizes that governance bottlenecks are often fundamentally technical, as we can't regulate what we can't measure, and desired practices can't become standard until they're cheap and replicable.