With his first encyclical, Pope Leo XIV has placed artificial intelligence at the center of his pontificate. The teaching document titled “Magnifica Humanitas” (“Magnificent Humanity”) addresses the technology’s impact on people and their coexistence. The text was presented to the public on Pentecost Monday at the Vatican in the presence of the Pope, comprising 105 pages in the original Italian.

The Pope’s Demands

Leo XIV, the first American to lead the Catholic world church, views AI as “a valuable tool that requires caution.” At the heart of his critique lies the question of power: control over the technology, he argues, does not rest in the hands of states but with “large economic and technological actors.” This poses a risk of opacity.

Specifically, according to Vatican News, the Pope calls for a vigilant engagement with AI in which responsibilities (accountability) remain clear at all times, appropriate policy and legal frameworks, independent oversight, and education of users. Above all, an ethics code is needed that follows the criteria of social justice. The encyclical states: “A more moral AI is of no use if this morality is determined by the few.”

The ecological dimension also finds its way into the document: “Current AI systems require large amounts of energy and water, have significant impacts on carbon dioxide emissions, and consume resources on a large scale.” Ahead of publication, Leo XIV had already emphasized that the challenge was “not technological, but anthropological in nature.”

The Position on Autonomous Weapons Systems

Particularly explicit – and unusually extensive – is the engagement with the military dimension of AI. Several sections of the document are dedicated explicitly to the question of AI and war, structured into dedicated chapters of the original text under the headings “The Normalization of War,” “Unbounded Violence,” and “Weapons and AI.”

Leo XIV bases his argument on the maxim “Artificial intelligence must be disarmed.” His point is explicitly not a blanket rejection of technology. “To disarm does not mean to reject technology, but to prevent it from dominating humanity,” he writes. Rather, AI must be freed from the “logic of military, economic, and cognitive competition.”

The core of his critique is the lowering of the threshold to war. The Pope cites an earlier statement by the Holy See: “The increasing ease with which operationally autonomous weapons systems can be deployed makes war ‘more feasible’ and subjects it less to human control, which contradicts the principle that the use of armed force should only occur as a last resort in cases of legitimate defense.” With this, Leo XIV ties into the Catholic concept of the “just war” and states that AI undermines this ethical corrective.

The Pope draws a clear red line on decisions about human life. Under no circumstances should machines alone decide over life and death, the document states. Human control over lethal decisions cannot be delegated – a point that aligns with the positions of many arms-control NGOs that have for years demanded a ban on “Lethal Autonomous Weapons Systems.”

Closely connected is the argument of moral distance: “Any technology that makes it easier to strike without seeing the face of the opponent lowers the moral threshold of conflict,” writes Leo XIV. The decisive sentence, which was quoted around the world on the day of publication, reads: “There is no algorithm that can make war morally acceptable.” Technology, he argues, can make wars “faster and more impersonal” – thereby lowering the moral threshold for violence.

The Pope places this development in a broader geopolitical context. He criticizes the expansion of the arms industry, the nuclear arms race, and the emergence of new armed groups – including jihadist movements – which further fuel conflicts out of interests of power. In the same breath, he also condemns the territorial logic of modern wars: “Any attempt or plan to extinguish or subjugate a nation is gravely immoral and therefore unacceptable.”

In response, Leo XIV calls for strict, internationally shared ethical guidelines and explicitly links the weapons question with the crisis of multilateralism, which he addresses in a dedicated section of the encyclical. At the presentation in the Vatican – a novelty, with the Pope himself in attendance – he urged vigilance in the face of rapid development.

Anthropic on the Vatican Stage

Remarkable is not only the content of the encyclical but also the setting of its presentation. Alongside curial cardinals, Chris Olah – co-founder of the AI company Anthropic and head of its interpretability team – also spoke. It was not the first time Olah had stood in a room filled with Catholics: rather, it was the latest step in a series of overtures by the multi-billion dollar AI company to religious leaders. In March, a small group of Christians had gathered at Anthropic’s San Francisco headquarters for a meeting partly organized by Olah.

Brian Patrick Green, a tech ethics expert at a Catholic university in Silicon Valley, was among several religious leaders, theologians, and ethicists who have participated since January in sometimes day-long conversations with the company – including sessions with the programmers who develop the AI models themselves. In April, Green attended a meeting between Anthropic programmers and around 15 Christian leaders.

According to Anthropic itself, this is part of a broader dialogue: in a press release, the company stated that “over the past several months” it had been organizing dialogues with groups whose work and traditions are relevant to the questions raised by AI. The first round of discussions had been held with “wisdom traditions” – with scholars, clergy, philosophers, and ethicists from more than 15 religious and cultural backgrounds.

The “Faith-AI Covenant” as an Interreligious Format

In parallel, a second format has established itself, in which competitor OpenAI also participates. Leading representatives of various religious groups recently met with representatives from Anthropic and OpenAI at the first “Faith-AI Covenant” roundtable in New York to discuss how morality and ethics can be embedded in the rapidly developing technology.

The meeting was organized by the Geneva-based Interfaith Alliance for Safer Communities, which addresses issues such as extremism, radicalization, and human trafficking. Further roundtables are planned in Beijing, Nairobi, and Abu Dhabi, among other locations. According to co-organizer Baroness Joanna Shields, the goal of the initiative is ultimately “a set of norms or principles” co-created by various groups and faiths – from Christians to Sikhs to Buddhists – which companies will commit to.

That finding a common denominator will be difficult is acknowledged by the initiative itself. “Religious communities see priorities differently,” said Rabbi Diana Gerson, a participant in the roundtable and associate executive vice president of the New York Board of Rabbis.

Historical Context and Criticism

Contacts between the Vatican and the tech industry are by no means new. Under Pope Francis, the Vatican initiated the Minerva Dialogues in 2016 – named after the Roman basilica Santa Maria sopra Minerva – which became annual discussions between church officials and tech leaders on AI ethics. In 2020, the Pontifical Academy for Life developed the “Rome Call for AI Ethics,” signed by Microsoft, IBM, and Cisco, among others. Anthropic is not among the signatories.

Not all observers view the close connection between tech companies and faith communities favorably. “I think most religious people, and certainly people from most Abrahamic faiths, would object to the idea that a system like Anthropic’s Claude could ever have personhood,” said Will Jones, who leads the faith outreach efforts at the Future of Life Institute.

The timing of the rapprochement is also interpreted differently. Some analysts describe Anthropic’s presence at the official presentation of the document as a savvy business move – the company, which is currently at odds with the Trump administration, is said to be trying to gain both moral and market footholds, particularly in European countries.