Global Shift Toward Inclusive AI Stewardship

Pope Leo XIV’s new encyclical Magnifica Humanitas calls for protecting people from concentrated AI power, defending jobs, and stronger oversight, framing AI as an ethical and social issue that religious leaders can influence. The pope’s engagement follows recent meetings between tech firms — including Anthropic, Meta, Google and Amazon — and religious authorities; Anthropic co-founder Christopher Olah was singled out. The piece highlights risks from AI systems reflecting dominant cultural or religious values, noting research showing bias toward Catholicism and policy responses such as Egypt’s ban on AI Quran interpretation. Experts warn AI customization will favor powerful cultures, raising questions about whose values shape global AI norms.

Pope Leo publicly criticized artificial intelligence, arguing that reliance on AI weakens human creativity and moral judgment. Speaking at a Vatican event, he warned technology can erode personal responsibility and the human capacities that define dignity, urging ethical reflection rather than outright rejection. The remarks target AI developers, tech companies, policymakers and faith communities, stressing the need for moral frameworks around deployment and use. While not proposing specific regulations, the pope’s statements add moral weight to ongoing debates about AI ethics, autonomy, and societal impact, potentially influencing public opinion and policy discussions in regions where the Vatican carries cultural authority.

EWTN News reports that the Pope urged Catholic priests to follow established “liturgical norms” to prevent confusion during the celebration of Mass. According to the brief item, the message emphasizes adherence to official rubrics and standardized practices rather than improvisation, with the stated goal of maintaining order and clarity in worship. The report identifies the Pope as the key figure and priests as the intended audience, but provides no additional details such as the date, location, specific document or speech, or examples of disputed practices. With limited information beyond the headline, the significance is framed as reinforcing uniformity in Catholic liturgy and reducing disorder in parish celebrations.

Author argues the main risk from AI isn’t sentient robots but concentrated commercial and political control as large investors push AI into everyday systems. They warn that wide adoption driven by venture capital and big tech could centralize power, automate surveillance and decision-making, and entrench business models that prioritize scale and profit over safety, fairness and public oversight. The piece calls for scrutiny of who builds and governs AI, stronger regulation, and alternative open or community-led approaches to prevent harms from opaque models and monopolistic ecosystems. It matters because AI’s design and deployment choices will shape economic power, civil liberties and digital infrastructure.

Analysts say parts of Pope Leo XIV’s new encyclical, Magnifica Humanitas — the Vatican’s first major letter focused on AI — may have been written with AI. Research posted on LessWrong by Linch Zhang and subsequent checks with the AI detector Pangram flagged between 40% and 100% of some paragraphs as AI-generated; The Verge’s own Pangram check estimated roughly 46% AI-written in a 2,000-word sample. The analysis cites stylistic markers linked to models like Anthropic’s Claude and contrasts sections Pangram labeled as clearly human. Experts caution that AI detectors aren’t infallible, though Pangram is widely used and reports a low false-positive rate. The Vatican has not yet responded to requests for comment.

4万字《壮丽人性》长文首发，教皇联手Anthropic，警告AI不能统治人类

Pope Leo XIV released Magnifica Humanitas, the Vatican’s first AI-focused encyclical, and publicly engaged with Anthropic co-founder Chris Olah, who urged scrutiny of AI by governments, religious institutions and civil society. The encyclical frames AI as a societal rupture comparable to the Industrial Revolution, warning that automatic prosperity is often illusory and highlighting labor displacement, dignity and distribution of benefits. The meeting signals a shift: some AI leaders now seek legitimacy and governance from traditional institutions rather than relying solely on market forces. This contrasts with U.S. policymaking, where recent moves—like President Trump withdrawing an AI executive order under industry pressure—show continued deference to commercial interests.

The Vatican-Anthropic relationship that's reshaping the AI ethics debate

The Vatican released Magnifica Humanitas, Pope Leo XIV’s encyclical on AI, framing the Church’s social teaching for the AI era and urging protection of human dignity, justice, and labour. The pope links his message to historical social encyclicals and calls for development that centers people rather than wealth. The document flags hard technical and ethical issues: it describes LLM interpretability limits, notes developers’ partial understanding of emergent systems, and warns about cultural bias, sycophancy, and the persuasive sheen of AI outputs. Techwriter Simon Willison highlights accessible prose and practical reflections, and notes tools used to consume the text (ElevenReader). The encyclical matters as a moral and policy-oriented voice shaping public debate on AI governance.

Pope Leo XIV published a 40,000-word encyclical, Magnifica Humanitas, urging the “disarmament” of AI and calling for technology to serve the common good rather than become an instrument of domination. With Anthropic’s co-founder present, the letter criticizes autonomous weapons, data colonialism, and concentration of control over patents, algorithms, platforms, and data, and updates Catholic social teaching for the AI era. The pope warns against new forms of extraction—especially health and demographic data—that can entrench power imbalances, while endorsing responsible uses of AI (e.g., the Vatican’s multilingual service translation tool). The encyclical frames a moral and policy agenda for equitable, human-centered technology governance.

The Vatican released Magnifica Humanitas, Pope Leo XIV’s encyclical on AI, framing AI as a social and moral challenge akin to the industrial revolution. The document highlights interpretability limits of large models, warns that AI systems are more “cultivated” than fully designed, and stresses human-centered development that protects dignity, justice, labor, and future generations. It calls out cultural bias and sycophancy in AI outputs and discusses socioeconomic risks from unequal access and externalized harms. The piece is notable for accessible prose and for positioning Catholic social teaching as a lens for tech policy and ethics, potentially influencing public discourse, policymakers, and faith communities engaging with AI governance.

Pope Leo XIV published a 40,000-word encyclical, Magnifica Humanitas, calling for the “disarmament” of AI to serve the common good and warning against AI’s use for domination, exclusion, and lethal autonomy. Presented in Rome with Anthropic’s co-founder in attendance, the pope singled out autonomous weapons, data neo-colonialism, and concentration of control—patents, algorithms, platforms, infrastructure, and data—as core concerns. He chose strong language to spark public debate and propose ethical and policy directions, framing AI governance as both moral and practical. The document is likely to influence tech policy discussions, corporate ethics, and international regulation debates around AI and data power.

Anthropic co-founder Chris Olah spoke at the Vatican for Pope Leo XIV’s May 25, 2026 encyclical “Magnifica humanitas,” arguing AI governance requires voices outside industry incentives. Olah warned that frontier AI labs face commercial, geopolitical, and personal pressures that can conflict with doing the right thing, and praised the Church’s role in ethical scrutiny. He emphasized AI systems are trained from human language and culture—“grown” rather than engineered—and said their character and societal impact are questions for humanities, religion, and public deliberation. Olah highlighted discernment on duties to the global poor, ethical design, and long-term risks as areas where broader engagement matters for tech’s future.

随着一份关于人工智能的宣言发布，教皇方济各十四世加入了那些呼吁世界变革的教皇行列

Pope Leo XIV issued an 83-page encyclical, Magnifica Humanitas, warning that AI concentrated in the hands of Big Tech and militaries risks widening inequality, weakening democracy and undermining human dignity. He urged that AI be "disarmed" from military and narrow economic interests, subject to stricter national and international rules, and opened to broad public participation in design, governance and benefits. The pope invoked the industrial-revolution-era precedent Rerum Novarum, stressed prioritizing human dignity over algorithmic power, and presented the document alongside theologians and Chris Olah of Anthropic, who backed broader societal oversight of AI development. The encyclical calls for regulation, inclusive governance and ethical stewardship to steer AI toward justice and solidarity.

Anthropic公司的奥拉表示，人工智能应由大型科技公司以外的机构来主导

Anthropic co-founder Chris Olah warned that AI development should not be left solely to tech companies, urging oversight from religious leaders, governments and civil society. Speaking at a Vatican event tied to Pope Leo XIV’s first AI encyclical, Olah said AI could displace large-scale human labor and that supporting displaced workers would be a moral imperative. He argued that commercial, geopolitical and personal pressures can conflict with the public interest inside leading AI labs, so external scrutiny and informed critics are needed. Olah highlighted three priority issues: mass unemployment risk, equitable global distribution of AI benefits, and the opacity and explainability of increasingly complex systems.

教宗利昂在其首份充满激情的通谕中呼吁全球在人工智能领域“踩下刹车”