I helped design the system that brought down ISIS financing. I’ve got an AI governance idea the Pope and Anthropic would both like

A Harvard fellow and former financial intelligence official proposes a new AI governance framework modeled after the international financial system used to combat ISIS financing, rather than nuclear arms control analogies. The approach emphasizes collaborative oversight mechanisms and transparency protocols that have proven effective in tracking illicit financial flows.

The proposal challenges the prevailing narrative in AI governance circles, which frequently draw parallels to Cold War nuclear deterrence models. Instead, the author advocates for a multilateral financial intelligence framework—specifically the mechanisms developed to disrupt terrorist financing networks. This model operates through information sharing, coordinated regulatory enforcement, and real-time transaction monitoring across jurisdictions, rather than mutually assured destruction logic.

The ISIS financing framework succeeded because it created enforceable accountability without requiring comprehensive bans on financial systems themselves. Banks, governments, and intelligence agencies coordinated to identify suspicious patterns and cut funding flows at source. Applied to AI, this suggests governance structures focused on monitoring model deployment, tracking training data provenance, and establishing transparency standards rather than attempting to constrain AI development globally through treaty-based limitations.

For the cryptocurrency and broader technology sectors, this governance model carries significant implications. Financial intelligence frameworks have historically adapted quickly to emerging technologies—from wire transfers to digital assets. An AI governance system modeled on these mechanisms would likely emphasize auditability, disclosure requirements, and multi-stakeholder oversight rather than heavy-handed restrictions. This approach could enable faster innovation while maintaining safety guardrails.

The model's resonance with both the Catholic Church (through papal dialogue) and organizations like Anthropic suggests potential consensus-building value. However, effectiveness depends on actual implementation and enforcement mechanisms. Success requires voluntary participation from major AI developers and international coordination—challenges that have historically hampered financial intelligence efforts in developing nations and jurisdictions with weaker institutional capacity.

• →AI governance should model financial intelligence frameworks rather than nuclear arms control treaties
• →The ISIS financing system demonstrated how coordinated monitoring and transparency can constrain harmful activities without banning underlying technologies
• →Multi-stakeholder oversight with information-sharing protocols may prove more effective than top-down regulatory restrictions
• →This approach potentially enables faster AI innovation while maintaining safety accountability measures
• →Success depends on voluntary participation from major AI developers and genuine international coordination