Pope Leo XIV will issue his first encyclical on May 26, focusing directly on AI, with Anthropic co-founder and Claude creator Chris Olah invited to share the stage. The Vatican simultaneously establishes an AI Commission. A 2,000-year-old institution is attempting to fill the AI governance vacuum with moral authority—covering a population larger than any AI bill's jurisdiction.
An institution with 1.4 billion followers, enduring for two millennia, has placed AI as its top agenda, even pulling in the creator of Claude to jointly release a position paper.
The core question this document aims to answer is precisely the one the entire AI industry cannot avoid—As machines become more and more like humans, what is human dignity and irreplaceability truly built upon?
On May 26, Pope Leo XIV will issue the encyclical "Magnifica Humanitas" ("Magnificent Humanity") at the Vatican, focusing on "protecting the human person in the age of AI."
An encyclical is one of the highest-level teaching documents of the Catholic Church, addressed to 1.4 billion Catholics worldwide, and also a symbolic action for a new Pope to highlight his governance priorities.
This Pope, an American with a mathematics background, has consistently been concerned with ethics in the AI age, aiming to protect humanity.
The Pope's First Encyclical, Written for AI
This release breaks two conventions.
Popes usually do not personally attend encyclical releases, delegating the task to a cardinal, but Leo XIV has chosen to be present.
Simultaneously inviting Anthropic co-founder Chris Olah as a lay spokesperson is almost unprecedented in the history of encyclical releases.
Vatican journalist Andrea Vreede's assessment is, "If only a cardinal released it, no one would really listen, but with the Pope present, all lenses will be focused there."
The timing was carefully chosen.
The encyclical was signed on May 15, coincidentally the same day 135 years ago when Pope Leo XIII signed the encyclical "Rerum Novarum."
Leo XIII
"Rerum Novarum" responded to the impact of the Industrial Revolution on labor rights and is regarded as a foundational document of Catholic social teaching.
Leo XIV explicitly referenced this connection in his inaugural address: "The Church offers the treasury of its social teaching to everyone in response to yet another industrial revolution, and to developments in the field of artificial intelligence."
Almost simultaneously, the Vatican approved the establishment of an AI Commission on May 16, integrating seven curial institutions for the first time to coordinate previously scattered AI efforts.
Previously, Leo XIV had already described the militarization of AI as pushing the world into a "spiral of destruction" in a speech at Europe's largest university.
Why Anthropic?
Olah's invitation is a precise alignment of mutual interests.
Olah leads interpretability research at Anthropic, aiming to open AI's black box and understand what neural networks are actually doing internally.
This happens to be the technical version of the Vatican's core concern: Can AI be fully understood and thus trusted?
Anthropic's engagement in religious ethics is systematic.
The company has previously invited clergy to participate in formulating Claude's "Constitution."
Further reading: Anthropic Officially Open Sources Claude's 'Soul'
In late April, representatives from Anthropic and OpenAI jointly participated in the inaugural "Faith-AI Covenant" roundtable in New York, discussing AI ethical frameworks with leaders from Hindu, Sikh, Jewish, Orthodox Christian, and other religious traditions. Follow-up meetings are planned in several cities globally.
A more pragmatic stake lies in Anthropic's current friction with the US government regarding the military use of AI, insisting its models cannot be used for autonomous weapon targeting and mass surveillance. This stance aligns highly with the Holy See's opposition to AI weaponization.
For Anthropic, the Vatican's moral endorsement strengthens its "responsible AI" brand positioning.
For the Vatican, having a company actually conducting AI safety research on stage elevates the encyclical beyond empty preaching. Both parties get what they need.
Silicon Valley long held skepticism towards organized religion.
Brian Boyd of the Future of Life Institute interprets this shift: "Silicon Valley's motto was 'move fast and break things.' They broke too many things and too many people."
Aligning with religious authority is an attempt to rebuild overdrawn credibility.
Ethical Outsourcing or PR Prop?
The narrative sounds warm, but critics see a different picture.
The sharpest criticism comes from Dylan Baker, Research Lead at the Distributed AI Research (DAIR) Institute.
He argues that discussions around "ethical AI" obscure a more fundamental question: Should certain AI systems be developed at all?
When the discussion framework becomes "since we're building it, let's build it well," the preliminary question of "whether to build it" is completely bypassed.
Humane Intelligence CEO Rumman Chowdhury, former AI Science Envoy for the Biden administration, is similarly lukewarm: "At best it's a distraction, at worst it's derailing the issues that really matter."
She believes Silicon Valley is trying to find "universal ethical principles" from religion to handle gray areas, but the value differences among global religious traditions are no smaller than regulatory differences between nations.
The structural dilemma is even more evident.
Papal encyclicals rely on moral suasion and have no legal binding force.
The Faith-AI Covenant is a voluntary pledge; participants can withdraw at any time.
The EU AI Act carries fines of up to €35 million or 7% of global revenue, but its extraterritorial reach is limited.
The US federal level still lacks a unified AI regulatory framework.
Currently, no single mechanism globally can effectively constrain the pace of AI development.
The Holy See is filling a vacuum of discourse, not a vacuum of governance capability.
As Machines Become More Like Humans, What Do Humans Rely on to Confirm They Are Human?
All discussions on AI ethics, whether rooted in religious doctrine, legal statutes, or corporate values, ultimately point to the same underlying question: Where does human irreplaceability lie?
People have offered many answers in the past: intelligence, creativity, linguistic ability, emotional resonance, moral judgment.
These answers are being shaken one by one.
AI can write poetry, compose music, pass the bar exam, and make users form real emotional attachments to it.
When these abilities once considered "what makes us human" can all be simulated, what can humans still use to define themselves?
The answer might lie in an attribute AI can never possess: finitude.
Humans die.
Precisely because life has an end, every choice carries weight, every sacrifice has a cost, every promise holds meaning.
Moral judgment is never just a product of logical deduction.
It's built on the visceral experience that "making a mistake causes pain, regret, and the loss of something irreplaceable."
AI can simulate the process of weighing pros and cons, but it lacks the capacity to bear consequences, the urgency brought by mortality, the existential anxiety of "having only one chance in this lifetime."
In the face of the advent of AGI or even ASI, the scenario truly worth vigilance is humans, while having the power of choice, actively outsourcing judgment, decision-making, and even moral responsibility to machines.
Humans are too eager to escape the act of "making a decision and then bearing the consequences," and AI happens to provide a seemingly perfect outlet.
The Pope's encyclical, Anthropic's Constitution, the Faith-AI Covenant roundtables—different in form, but sharing the same underlying message: Technology can evolve infinitely, but the one who ultimately presses the button must be a mortal, vulnerable human being who feels pain and must be responsible for the consequences.
This is likely what "Magnifica Humanitas" truly wants to say.
References:
https://www.theguardian.com/world/2026/may/18/pope-leo-encyclical-human-dignity-ai-anthropic
This article is from the WeChat public account "Xinzhiyuan," author: Xinzhiyuan
