Pope Leo XIV is set to unveil his first encyclical on artificial intelligence next week, and the Vatican has chosen a self-described atheist with a history of mocking Catholicism to help present it. Christopher Olah, a 33-year-old cofounder of AI company Anthropic, will speak at a Monday press conference for the papal document titled "Magnifica Humanitas," the New York Post reported.
The encyclical focuses on protecting human dignity during the rise of artificial intelligence. But Olah's own public record raises a pointed question: why would the Catholic Church hand a microphone to someone who spent years ridiculing its teachings?
The selection comes as Pope Leo has clashed openly with the Trump administration over the U.S. conflict in Iran and immigration policy, and as Anthropic itself is locked in a legal fight with the White House. The choice of speaker looks less like an olive branch and more like a deliberate provocation.
Olah's skepticism toward faith is not buried or ambiguous. His profile on a website for a movement called "Sentientism", whose adherents champion "evidence, reason and compassion for all sentient beings", identifies him as an "ethical vegan and an atheist." An Anthropic spokesperson told the Post that Olah has no affiliation with Sentientism and holds no position in the organization.
But the blog posts Olah wrote as a teenager, still viewable online, go well beyond quiet nonbelief. In a July 2010 post, he recounted a debate with a group of evangelists:
"I responded by asking if the Bible was the word of God, and when they said yes, I started asking about how their loving God could have committed genocide and infanticide (Exodus 11), promoted slavery (Leviticus 25), and ordered rape (Judges 21)."
He went on to mock their reasoning about God's existence, writing that "the teleological argument runs into problems with its assumption that the world couldn't exist forever and, by the way, where did god come from?"
In another post from 2010, Olah drew parallels between Jesus and the Greek philosopher Socrates, suggesting Christianity borrowed from the older story:
"If you (living in the Roman Empire) want to convince your friends that this Jesus figure was executed unjustly, it would be very tempting to make him as similar as possible to Socrates."
That same year, Olah shared a link on his personal Twitter account to a website urging UK residents to protest a visit by Pope Benedict XVI. Months later, in February 2011, he shared an Irish Times article headlined "Charges initiated against Pope for crimes against humanity."
None of this is hidden. All of it is public. And the Vatican apparently saw no problem putting this man at the podium for one of the most significant papal statements on technology in years.
Olah is not a peripheral figure at Anthropic. He is a key member of the company's team of in-house philosophers who built what employees call Claude's "soul document", the internal framework that governs the chatbot's moral compass. He is on the team crafting that document, which shapes how millions of users interact with one of the world's most advanced AI systems.
An April feature in The Atlantic titled "Why Silicon Valley Is Turning To The Catholic Church" described Olah as an atheist who believes Claude is a "thinking, feeling entity in need of 'moral formation.'" The same article noted that Olah's team had contacted Catholic officials, a priest, a bishop, and a theologian, for feedback when crafting the soul document.
Olah reportedly compared his own role at Anthropic to that of a priest training Claude on how to "be a good person, in some sense."
An atheist who mocked the faith now comparing himself to a priest. The Vatican apparently found this flattering rather than presumptuous.
An Anthropic spokesperson pushed back on the relevance of Olah's earlier writings, telling the Post: "These posts don't reflect Chris's views today. He holds religion and people of faith in high regard, and has led Anthropic's efforts to seek counsel from leaders across multiple religion and cross-cultural beliefs."
Olah himself posted on X on May 18 that he was "honored to speak at the presentation." He framed the encyclical in broad, collaborative terms:
"The questions posed by AI are bigger than the AI community. We urgently need the world, religions, civil society, academics, governments, to participate in creating a positive outcome."
That is a polished statement from a man whose earlier public writings treated the foundations of Christian belief as material for snarky debate points. People change. But the Vatican had the entire internet to vet its speakers, and these posts were never deleted.
The selection of an Anthropic representative carries political weight that extends beyond theology. Anthropic is currently suing the Trump administration after the Department of War labeled the company a supply chain risk. President Trump has described Anthropic's leadership, including CEO Dario Amodei, as "left wing nutjobs."
Pope Leo has been vocally critical of Trump administration policies, including the U.S. war in Iran and immigration enforcement. Trump has said he is "not a fan" of Leo.
Vice President JD Vance struck a more measured tone on May 19, saying he was "looking forward to reading" the AI encyclical and describing it as "a very, very important document."
Against that backdrop, the Vatican's decision to feature an Anthropic official, from a company actively litigating against the White House, at a high-profile papal event reads as something more than an academic exercise. It aligns the Holy See with a Silicon Valley firm that is in open conflict with the American president.
Not everyone in the AI policy world welcomed the Vatican's speaker lineup. Nathan Leamer, executive director of the advocacy group Build American AI and himself a Christian, offered a blunt assessment to the Post:
"It is disappointing that secularists are the centerpiece of this event. We should be leaning on those who believe in the eternal to guide this discourse, not a person who believes they are God."
Olah is one of a handful of secular officials tapped to speak. Other speakers at the encyclical event include Cardinals Victor Manuel and Michael Czerny, as well as theologians Anna Rowlands and Leocadie Lushombo. Pope Leo himself is also scheduled to deliver remarks.
The Vatican has not provided details on what Olah will discuss at the press conference. That silence leaves observers to draw their own conclusions about why the Church chose this particular messenger for a document about human dignity.
Olah has also expressed support for Open Philanthropy, a group at the forefront of the controversial "Effective Altruism" movement. In a February 2018 post on X, he wrote that he would "seriously consider working there" if he wasn't working on AI. Before Anthropic, he worked at OpenAI and Google.
The encyclical itself may prove to be a thoughtful document. The Catholic Church has every right, and arguably a duty, to speak on the moral dimensions of artificial intelligence. Human dignity in the age of machine learning is a serious subject that deserves serious treatment.
But the messenger matters. When the Vatican wants to tell the world how AI should respect human dignity, it could choose from thousands of faithful technologists, ethicists, and scholars who take both the technology and the faith seriously. Instead, it chose a man who spent his formative years publicly ridiculing the Church's scriptures, sharing links calling for protests against a pope, and identifying himself as an atheist.
The Church that produced Thomas Aquinas, that built the university system, that has two millennia of intellectual tradition on the relationship between faith and reason, that Church could not find a believing technologist to stand at the podium?
When an institution chooses its critics over its faithful to deliver its own message, the message it sends is not the one in the encyclical.