The framework is a starting point. It helps organize the topics that matter in AI governance so we don’t talk past each other. Think of it as a map: it doesn’t prescribe solutions, but it shows the terrain. Once we have a shared structure, it becomes much easier to discuss ideas, compare proposals, or identify gaps.
The pillars represent the major buckets of issues Congress will have to face as AI systems grow more powerful. Instead of treating “AI governance” as one giant blob, the pillars break it into 13 digestible areas like safety, elections, jobs, deepfakes, privacy, and national security.
If policymakers write new AI laws, each law will fall under one or more of the pillars. For example, a deepfake-labeling rule belongs under Pillar 9, while worker protections belong under Pillar 6. The framework doesn’t force laws into boxes—it simply helps people see how the pieces relate.
No. It’s an informal, community-friendly way to help people think about the structure of AI governance. Policymakers could build something similar or take it in a completely different direction. The purpose is to give regular people a shared language for discussing the issues.
Thirteen turned out to be the smallest number that still captured all major governance topics without lumping too much together. Fewer pillars gets too vague; more gets too complicated. It’s a practical middle ground. And the number isn’t locked in—if the group evolves the framework, it can change.
No. The framework doesn’t take positions on what policies should be passed. It simply organizes the topics so people with different political views can discuss the same issues clearly and respectfully.
The goal is to help regular people stay ahead of major AI changes and have a voice before the rules are written without them. A small, informed group can have an outsized impact. This framework is one of the tools to make that possible.