Pope Leo XIV Critiques AI: Ethics and Monopolies Under Scrutiny

Pope Leo XIV Addresses AI in an Unprecedented Encyclical

One year after his accession to the papal throne, Pope Leo XIV has published an encyclical titled Magnifica Humanitas, dedicated to artificial intelligence. This 50-page document represents a strong stance from the Vatican on issues such as digital monopolies and autonomous weapons.

This text, both theological and technical, comes on the occasion of the 135th anniversary of Rerum novarum, the founding encyclical of the Catholic Church's social doctrine. Magnifica Humanitas candidly examines the potential pitfalls of AI, from the concentration of data in a few hands to autonomous weapon systems. By transcending Catholic circles, this encyclical raises questions that the tech industry often prefers to avoid, especially after a year marked by notable AI missteps.

AI is Not Morally Neutral, According to Leo XIV

In his encyclical, Leo XIV rejects the idea that technology can be morally neutral. "We cannot consider AI as morally neutral," he asserts. According to him, every algorithm is imbued with human choices regarding what it should measure, favor, or ignore. The values of the designers are embedded in the code long before its use. Thus, questioning whether AI is used well or poorly comes too late in the reasoning process.

The text highlights a reality often overlooked by the industry: a modern AI is not "programmed" in the traditional sense. It is trained on billions of data points, and its creators cannot always explain its decisions. Leo XIV emphasizes that AIs are more "cultivated than constructed." This lack of clarity poses a major ethical problem: if no one truly understands the internal workings of the machine, who can be held accountable for its mistakes?

To illustrate his point, the pope uses two well-known biblical narratives. The Tower of Babel, a symbol of human arrogance, collapses into confusion. In contrast, the reconstruction of Jerusalem by Nehemiah is a humble, collective project undertaken for the good of all. "The choice is not between a 'yes' or 'no' to technology, but between building Babel or rebuilding Jerusalem." It is within this framework that the pope invites reflection on AI.

Digital Monopolies and Autonomous Weapons in the Pope's Crosshairs

On the issue of the concentration of digital power, the pope adopts a cutting tone. He criticizes the fact that patents, algorithms, platforms, and data are controlled by a small number of private companies. For Leo XIV, these resources should be accessible to all, as they represent "new forms of property" that must benefit all of humanity. He implicitly critiques the Sino-American bloc. "A more moral AI is useless if that morality is decided by a handful of people," he emphasizes. In other words, allowing Google, Meta, or OpenAI to define the ethical rules of AI alone would be akin to handing the keys of the house to those who have a vested interest in controlling access.

Regarding autonomous weapon systems, the pope is unequivocal. "There is no algorithm capable of making war morally acceptable," he states in the encyclical. This declaration reflects a concrete concern: when it is a machine that decides to fire, rather than a human being, no one truly assumes responsibility for the act. War becomes easier to initiate, victims are reduced to dots on a screen, and in the event of an error, it is impossible to identify a culprit. For Leo XIV, the decision to resort to lethal force must remain in the hands of identifiable humans who can be held accountable.

The text also delves into the working conditions of millions of workers, often in poor countries, who sort and label data for AI models. Without them, there might be no ChatGPT or Gemini. Before that, children extract cobalt or lithium in dangerous mines, essential for electronic chips. "Nothing in the world of AI is immaterial or magical," adds Leo XIV. What the pope expresses here is something rarely articulated so clearly by someone so visible.

For the Pope, AI Should be "Disarmed"

On the subject of transhumanism, which aspires to enhance humanity through technology, even merging it with machines, the pope remains firm. An AI, he reminds us, has no body, feels nothing, has never suffered or loved. "So-called artificial intelligences do not live experiences, do not possess bodies, and do not know joy or pain." Beyond the technical aspect, it is the underlying philosophy that worries him. Once we start to see humans as "material to be perfected," we inevitably create categories of the optimized and the others, the efficient and the useless. A logic that, pushed to the extreme, can justify that some lives count less than others.

Leo XIV proposes concrete measures, some of which have already been considered, such as banning access to social networks below a certain age, better training teachers in digital tools, and protecting minors' personal data. But the pope goes further, highlighting the violence of online language, amplified by algorithms that favor outrage and confrontation because they generate more clicks. "Let us disarm words, and we will contribute to disarming the Earth (...), regulating AI is not enough: it must be disarmed and made accessible."

One question remains that the text obviously does not address. What is the value of the Vatican's opinion against companies whose budgets exceed those of many states? The encyclical diagnoses the right issues, names the right pitfalls, but says little about how to truly correct them and by whom. It is significant, of course, to hear such a strong voice speak about algorithmic monopolies and digital slavery at a time when few leaders dare to do so for fear of upsetting investments. But a beautiful moral discourse, however just, is not enough to regulate OpenAI or make its competitors bend. Between stating the issues and having the means to change them, the gap remains wide.