A blueprint for democratic governance of frontier AI

The U.S. is at a critical juncture for establishing effective governance over advanced artificial intelligence. A new blueprint outlines a comprehensive strategy for building a durable federal framework to ensure the safety and responsible development of increasingly capable AI systems. This initiative aims to address the significant national security and public safety challenges posed by frontier AI.

The proposed framework is built upon three core pillars. First, it seeks to leverage the emerging consensus found in state-level frontier AI safety laws, creating a cohesive national approach. This includes drawing lessons from initiatives such as California’s SB 53, New York’s RAISE Act, and Illinois’s SB 315.

Second, the blueprint emphasizes strengthening CAISI, positioning it as the primary federal institution responsible for frontier AI safety. This dedicated focus will help centralize expertise and coordinate efforts.

Finally, the strategy calls for mobilizing a broader government-wide resilience plan. This comprehensive approach will ensure that various federal agencies are equipped to understand, manage, and mitigate the risks associated with cutting-edge AI technologies.

This federal effort builds on recent advancements, including the White House’s executive order on Promoting Advanced Artificial Intelligence Innovation and Security. By establishing a durable and adaptable framework, the U.S. can evolve its regulatory approach alongside the rapid advancements in AI technology.