The authors suggest the creation of three AI governance regimes: a global scientific body for fact-finding and advising on AI risks, a mechanism to manage tensions between major AI powers and prevent dangerous proliferation, and a body to maintain geopolitical stability amid AI-driven change. They argue that these regimes must be universally accepted, specialized, and dynamic enough to keep up with AI's evolution. The authors conclude that successfully governing AI will not only mitigate societal risks and ease geopolitical tensions but also establish a model for governing other disruptive technologies.
Key takeaways:
- The rapid advancement of AI technology is reshaping the global power structure, with tech companies wielding significant influence over their AI creations, potentially surpassing the control of nation-states.
- Existing regulatory efforts are inadequate to govern the evolving nature of AI, and a new governance framework is needed that is precautionary, agile, inclusive, impermeable, and targeted.
- Three AI governance regimes are proposed: a global scientific body for fact-finding, a mechanism to manage tensions between major AI powers and prevent proliferation of dangerous AI systems, and a body to maintain geopolitical stability amid AI-driven change.
- Successfully governing AI will not only mitigate societal risks and ease geopolitical tensions, but also establish a model for addressing other disruptive, emerging technologies in the future.