The Vatican has promoted an international declaration that proposes limiting the application of artificial intelligence in the military sphere and advancing toward nuclear disarmament. The document, titled Declaration of Rome for an Unarmed and Disarming Peace in the Age of Artificial Intelligence, Nuclear Weapons and the New Digital Protocols, was signed on July 16 at the Capitol in Rome at the conclusion of the Global Assembly of Nobel Laureates on Artificial Intelligence and Nuclear War, held over two days at Borgo Laudato si’.
The initiative, organized within the Holy See’s environment, brought together Nobel laureates, artificial intelligence specialists, former heads of state and government, and representatives of universities and research centers. The text expressly includes the call of Leo XIV to work for an “unarmed and disarming peace” and supports several of the proposals put forward by the Pontiff in the encyclical Magnifica humanitas on the ethical governance of new technologies.
A proposal to regulate the military use of artificial intelligence
The declaration maintains that humanity is going through “a decisive moment in its history” due to the rapid development of artificial intelligence and its impact on the economic, military, and social spheres.
The document acknowledges the opportunities offered by this technology, but also warns of the risks arising from its application in weapons systems, the increase in competition among nuclear powers, and the concentration of technological development in a limited number of states and large corporations.
In this context, the signatories consider it necessary to establish international mechanisms to regulate the use of artificial intelligence in strategic and military domains.
A treaty to maintain human control over nuclear weapons
One of the main points of the declaration is the proposal to negotiate an international treaty that prohibits the integration of artificial intelligence systems into the command, control, and launch mechanisms of nuclear weapons without effective human oversight.
The text states that “the final decision to employ a nuclear weapon must never be entrusted to an automated system” and urges governments to ensure that these decisions always remain under human responsibility.
It also proposes strengthening oversight of fully automated processes and preventing the malicious use of artificial intelligence in cyber operations targeting nuclear infrastructure.
The encyclical Magnifica humanitas, among the document’s references
The declaration expresses its support for initiatives inspired by the encyclical Magnifica humanitas of Leo XIV and backs the United Nations Independent International Scientific Panel on Artificial Intelligence.
Among its proposals is also the creation of a “digital commons” (digital commons), conceived as a framework to facilitate the exchange of data and knowledge that contributes to addressing challenges related to artificial intelligence, nuclear weapons, and other technological risks of global scope.
Call to advance nuclear disarmament
In its final section, the document requests the start of negotiations to achieve the verifiable and irreversible elimination of nuclear weapons and recalls the commitments undertaken by states in the Treaty on the Non-Proliferation of Nuclear Weapons (NPT) and the Treaty on the Prohibition of Nuclear Weapons (TPNW).
The declaration also urges nuclear-weapon states to adopt policies aimed at progressively reducing the role of these weapons in their defense strategies, with the goal of lowering the risk of nuclear conflict and strengthening strategic stability.