Pope Leo XIV's encyclical calls for global AI disarmament and regulation

A hypothetical encyclical from a non-existent Pope Leo XIV would dramatically escalate the Vatican's engagement with artificial intelligence. Unlike the 2020 Rome Call for AI Ethics—a voluntary, multi-stakeholder pledge signed by Microsoft, IBM, and other tech giants—this new document demands strict limits on military and commercial AI, including a binding global disarmament of autonomous lethal systems. The timing aligns with heightened debate around the EU AI Act's finalization in 2024 and ongoing United Nations discussions on lethal autonomous weapons, reflecting a broader societal shift from asking 'how to make AI ethical' to 'how to stop it from being weaponized.' The inclusion of Anthropic's co-founder at the Vatican presentation suggests an attempt to bridge religious ethics with the tech industry's safety-conscious fringe, but the encyclical's unilateral moral absolutism stands in stark contrast to the incremental, principle-based approach of previous initiatives.

The landscape of AI governance is fragmented. The EU AI Act provides enforceable rules based on risk categories but does not explicitly call for disarmament or challenge the fundamental legitimacy of autonomous weapons. Defense contractors like Palantir and L3Harris operate in a market for autonomous systems projected to reach $12 billion by 2027, a figure that would shrink under strict regulation. Commercial AI leaders such as OpenAI, Google, and Microsoft face increased compliance costs, potentially reducing profit margins. Meanwhile, Anthropic's focus on technical safety and interpretability positions it as a responsible leader, but the company still operates within the very commercial race the encyclical condemns. The Rome Call signatories represent a broad coalition, but their voluntary commitments have not halted military AI development in nations like the United States, China, and Russia.

The implications reveal a deep tension between moral urgency and practical enforceability. The encyclical adds ethical pressure that could influence investor sentiment, potentially driving capital toward ethical AI firms while raising risk premiums for defense contractors. However, the lack of enforcement mechanisms means developers may acknowledge the document but continue profit-driven work. The focus on disarmament risks overshadowing beneficial AI applications in healthcare, climate science, and economic development. Moreover, the encyclical does not address technical challenges such as defining 'autonomous lethal systems' or verifying compliance—issues that have stymied international treaties for years. The narrative also ignores that many AI risks—algorithmic bias, misinformation, labor displacement—are not directly related to weaponization. By framing AI primarily as a military threat, the encyclical may inadvertently narrow the regulatory agenda.

At bottom, this hypothetical encyclical is a symptom of a growing demand for global governance of AI, but it underscores the gap between moral leadership and binding treaties. The real story is not the Vatican's stance but the structural difficulty of translating ethical outrage into enforceable rules. As Yoshua Bengio has argued, moral authority alone cannot stop an arms race; verifiable compliance mechanisms are essential. The encyclical may rally civil society and pressure policymakers, but its ultimate impact depends on whether nations can agree on definitions, verification, and consequences. For now, the call for AI disarmament remains a powerful moral test—one that reveals how far we are from aligning the economics of warfare with the ethics of peace.