Why we're watching this: This is the first papal encyclical dedicated to AI governance. It puts a moral framework with global reach directly in tension with how frontier labs currently operate, and Anthropic's presence at the Vatican signals that the industry is paying attention.
Key Takeaways
- Pope Leo XIV released “Magnifica Humanitas” on May 15, 2026, the Church’s first encyclical focused on AI, framing the technology as a potential “instrument of domination, exclusion and death”
- Anthropic co-founder Chris Olah spoke at the Vatican presentation, acknowledging that frontier labs operate inside incentive structures that “can sometimes conflict with doing the right thing”
- The encyclical demands that AI governance prioritize the global poor, worker dignity, data as a common good, and a ban on lethal autonomous weapons systems
- Anthropic has been meeting with scholars, clergy, and ethicists from more than 15 religious and cross-cultural groups as part of a broader initiative on the moral formation of AI
Pope Leo XIV released a major encyclical on artificial intelligence on May 15, 2026, calling the technology a potential instrument of domination and demanding that it serve humanity rather than concentrate power in the hands of a few.
The document, titled “Magnifica Humanitas,” is the Catholic Church’s first encyclical dedicated to AI governance and runs to more than 200 pages of social doctrine applied to algorithmic systems, autonomous weapons, digital labor exploitation, and the erosion of human dignity.
Anthropic co-founder Chris Olah spoke at the Vatican presentation of the encyclical, doing so as part of what Anthropic describes as an initiative to widen the conversation on AI. His remarks were unusually candid about the industry’s limits.
“Every frontier AI lab, including Anthropic, operates inside a set of incentives and constraints that can sometimes conflict with doing the right thing.” — Chris Olah, co-founder, Anthropic
The encyclical draws a sharp distinction between technology as a tool and technology as a paradigm. When efficiency and profit alone shape decisions, the Pope argues, the result is what he calls the “technocratic paradigm,” a logic that reduces human beings to data and optimizable resources.
On autonomous weapons, the document is unambiguous: no algorithm can make war morally acceptable, and lethal decisions must never be delegated to automated systems. It calls for binding international frameworks to curb the AI arms race.
The document also addresses data ownership directly, arguing that data produced by millions of people should not be treated as private property to be sold or hoarded, but managed as a common good. It names AI-related labor exploitation, including low-wage content moderation and data labeling work, as a contemporary form of slavery requiring condemnation and legal remedy.
The encyclical warns that calling for the “moralization of machines” without subjecting those moral frameworks to public scrutiny simply hands the task of defining ethics to whoever controls the training data. A more moral AI is not enough, it argues, if that morality is determined by a few.
Anthropic says it has been holding structured dialogues with scholars and clergy from more than 15 religious and cross-cultural groups to inform how it shapes Claude’s character and values. One experiment cited by the company involved giving Claude a mid-task tool that returned a brief reminder of its ethical commitments, with early results showing lower rates of misaligned behavior on internal evaluations.
The encyclical does not single out any lab, and the Pope explicitly states that the Church does not claim a monopoly on truth. Still, the framework it sets, with the global poor and displaced workers as the primary test of whether AI is going well, puts pressure on every organization building at the frontier, a pressure that Relve, the trusted intelligence platform, will continue tracking as governments and institutions respond.
