Generated by Codex with GPT-5

This summary covers The Economist’s May 30th, 2026 Europe article listed in the contents as The pope and AI and published under the headline Of God and Claude.

Pope Leo’s first encyclical, Magnifica Humanitas, is presented as much more than a routine Vatican statement on modern life. The Economist reads it as an attempt to open a major philosophical fight with Silicon Valley’s most expansive vision of artificial intelligence. The document ranges widely, touching journalism, diplomacy, slavery and just-war doctrine. But the article argues that its central target is the unregulated development of AI and, more deeply, the idea that technology should remake or surpass human beings.

The immediate policy message is familiar enough. Leo wants artificial intelligence constrained, or in his phrase “disarmed”. That puts the Vatican close to European regulators who have been trying to control a technology whose builders do not fully understand its behavior. The article notes the presence at the encyclical’s launch of Chris Olah, a co-founder of Anthropic, who warned that researchers are encountering aspects of powerful models that remain mysterious and disturbing. In that setting the pope’s intervention is not framed as anti-modern scolding. It is framed as a warning that even insiders are unsure how far their systems can be trusted.

The deeper target

The article’s sharper point is that Leo is not only asking governments to write better AI rules. He is challenging two ideas that have become influential in parts of the technology world: transhumanism and posthumanism.

Transhumanism imagines the enhancement of people through tools such as body engineering, devices and algorithms. Posthumanism goes further, imagining a merger of humans, machines and the surrounding environment that could eventually carry humanity into a new evolutionary stage. The Economist connects this to the tech world’s language of the Singularity, the hypothesised moment when technological progress produces forms of intelligence and capability far beyond ordinary human limits.

Leo’s objection is not simply that these ideas sound strange. It is that they treat humanity as a defective thing to be optimised. The encyclical argues that this desire to perfect or surpass the human species can become morally dangerous, because it encourages people to sort lives into more useful and less useful categories. The article highlights the pope’s comparison with eugenics, the 20th-century ideology that promised biological improvement and helped justify monstrous politics.

Against that, Leo defends human limitation. Frailty, failure, scars and dependence are not bugs in the human condition. They are part of what makes moral and spiritual life possible. In the article’s reading, the encyclical is insisting that dignity cannot depend on upgradeability. A person does not become more worthy by becoming more efficient, less vulnerable or more machine-like.

A rival eschatology

The Economist treats the encyclical as the first shot in a philosophical turf war. Some AI optimism has the shape of a secular religion: a coming transformation, a new intelligence, a path to god-like capacities. That story resembles Christian expectation while replacing salvation with technical acceleration. The pope’s challenge is therefore not only legal or practical. It is theological. He rejects the idea that artificial general intelligence should be treated as a messianic event.

That makes the politics awkward. Some prominent AI enthusiasts on the American right, including influential investors, are also vocal Christians. The debate is not a simple contest between religion and technology, or between Europe and America. It is a dispute inside a wider culture that wants both spiritual language and technological transcendence.

The article is careful to say that Leo does not reject technology itself. He acknowledges that technological development has improved human life. The argument is about limits, purpose and power. Technology is acceptable when it serves human flourishing. It becomes dangerous when it turns human beings into raw material for optimisation or treats their replacement as progress.

The piece ends with an irony. Modern papal encyclicals are not written entirely by the pope; aides draft routine passages. A tech blogger’s analysis suggested that parts of Magnifica Humanitas may themselves have been written by AI, specifically Claude, because of stylistic patterns associated with Anthropic’s model. One detector flagged part of the opening as AI-generated, whereas previous encyclicals showed no such signal.

That twist does not undo Leo’s argument. It makes the argument more concrete. AI is already slipping into institutions that want to judge it, including old and self-consciously human ones. The real question is not whether the Vatican, governments or companies can keep AI outside the gates. It is whether they can use it without accepting the worldview that treats human beings as merely the next platform to be improved, merged or replaced.