Artificial Intelligence and the Catholic Church: 3 Questions for Discernment in a Speech by the Co-founder of Anthropic to Pope Leo XIV
Full text in Spanish of the speech by Christopher Olah, co-founder of the AI company Anthropic, to Pope Leo XIV on the encyclical Magnificas Humanitas
4 months ago
The co‑founder of Anthropic delivered a speech to Pope Leo XIV in which three questions were posed to guide the Catholic Church’s discernment of artificial intelligence 1.
The article notes that the speaker framed three distinct questions for the Pope to consider regarding AI’s role, ethical implications, and pastoral guidance within the Church 1.
The questions were intended to prompt thoughtful reflection and discernment by Church leadership on how emerging AI technologies intersect with Catholic doctrine and mission 1.*
Examine Catholic ethical guidelines for artificial intelligence development
Catholic ethical reflection on artificial intelligence (AI) treats it neither as a moral “neutral” tool nor as a new form of “human agency,” but as a powerful instrument whose moral value depends on human purposes, governance, and oversight. Across recent Vatican teaching and episcopal guidance, several recurring guidelines emerge for AI development: human dignity at the center, the common good as the aim, protection of truth, inclusion of the poor and vulnerable, transparency/accountability, human control of decisions, and responsible care for creation and peace.
Catholic ethics begins with an anthropology: intelligence is a gift connected to being created “in the image of God”, and technology must serve the vocation of the human person—not displace it. The Church explicitly frames AI within this stewardship: humans are “co-workers in the work of creation”, called not merely to consume outputs but to act responsibly in light of what makes human beings human (freedom, love, genuine relationships, openness to truth and beauty).
A key ethical warning follows: AI must be understood as qualitatively distant from human prerogatives of knowledge and moral action, even when it “simulates” capabilities. Francis stresses that such devices lack the human quality of conscience and moral autonomy, so they can become socially dangerous when their simulated capacities are mistakenly treated as if they were human judgment.
Catholic teaching repeatedly insists that AI’s effects depend not only on engineering, but on the aims and interests of owners and developers and on how it is used in particular contexts. Francis notes that we cannot assume AI will benefit humanity or peace unless development is guided by fundamental values.
This yields a practical moral criterion: regulation and development should foster an “ethic inspired by a vision of the common good,” described as an ethic of freedom, responsibility, and fraternity, supporting the full development of people in relation to others and to all creation.
A central guideline is that AI must not become a hidden substitute for human moral agency in areas where persons are affected.
Francis, speaking on AI and decision-making, emphasizes that “decision-making … must always be left to the human person.” He warns that removing a person’s ability to decide about their own life dooms humanity to a future “without hope.”
Related to this, Francis describes ethically risky uses of AI systems that “help judges” predict future conduct. The moral danger is that such systems can implicitly delegate to a machine what should be a human discernment about a person’s future, while also embedding prejudices inside categories and data. He specifically points out how features like ethnic classification or prior minor infractions can influence outcomes—and notes that human beings are capable of change in ways a machine cannot fully account for.
Catholic ethics therefore treats human oversight and accountability as not optional add-ons but as ethically required structures. The USCCB underscores that AI systems must have human oversight and well-defined accountability to promote transparency and fair democratic processes.
A recurring “test of legitimacy” in Catholic guidance is how AI affects the vulnerable. The Church and bishops connect AI ethics to the preferential concern for the poor and to social justice.
Francis states that the positive outcome of AI development depends on safeguarding values such as inclusion, transparency, security, equity, privacy, and reliability, and on resisting the temptations of selfishness, self-interest, profit-seeking, and the thirst for power.
The USCCB further stresses that AI will “only serve all” when it works to assist the poorest and most vulnerable, enabling them to participate equitably in development and use. It explicitly warns that without safeguards, AI could perpetuate marginalization and discrimination, widen the digital divide, and worsen inequalities.
The Catholic note also clarifies that “bias” in this context includes systematic algorithmic errors and bias introduced by biased training data—an important technical concept for moral governance, because it explains why “neutral” datasets can still produce unjust outcomes.
Catholic ethics pays special attention to information integrity, since AI reshapes how people communicate and form judgment.
The USCCB references a “growing crisis of truth” and warns about the use of AI—especially via deepfakes and misinformation—to manipulate public opinion and undermine respect for the truth and for persons’ dignity. The moral requirement again is human oversight and accountability to protect transparency and fair democratic processes.
Francis also ties this to the broader human question: AI debates must not reduce the human world to what can be expressed only in numbers and predetermined categories, excluding other forms of truth.
While Catholic documents do not provide a software engineering checklist, they repeatedly name governance values that must be protected in practice.
Francis explicitly lists inclusion, transparency, security, equity, privacy, and reliability as values that must be respected for AI to contribute positively to humanity and peace.
The USCCB likewise links ethical principles to policy mechanisms requiring transparency and accountability (especially in systems affecting democratic and civic life).
Catholic ethics treats work as inherently ordered to human dignity and the good of the person—not merely as an economic variable.
The USCCB notes that “labor (work) has inherent worth” and is essential to a person’s dignity, quoting the Church’s social teaching that human work proceeds from the person and is ordered to the human person as its final goal.
Accordingly, AI’s labor-saving benefits must not deprive people of dignity through job displacement, inequality, or exploitation. The US bishops urge protections for workers, education and job training, and safeguards that include public accountability and human oversight in AI-driven employment decisions.
Catholic ethical guidance applies to the social and moral ecology where AI is used.
The USCCB calls AI policy to strengthen and support family life and counteract technology’s isolating effects.
It also explicitly raises moral protections for children online and against pornography, including “virtual child pornography.”
Catholic ethics is explicitly skeptical about weaponizing AI in a way that undermines human control.
The US bishops state that they have “consistently raised concerns” about lethal autonomous weapons and cite the moral principle that war’s destructive power over innocents becomes uncontrollable. They therefore call for policies making clear that human control over any weapon system is essential to mitigate the horrors of warfare and protect fundamental human rights.
AI ethics also includes ecological responsibility.
The US bishops urge that AI development and policy remember both “the cry of the earth” and “the cry of the poor.”
They highlight concrete harms tied to AI infrastructure, including significant energy expenditures and the production of electronic waste containing harmful substances. They ask for policies promoting energy efficiency and responsible recycling of resources/components.
Catholic teaching does not stop at product-level ethics; it requires institutional responsibility and education for those shaping AI’s trajectory.
Francis argues that the expansion of technology needs accompanying formation in responsibility and that there is a need to strengthen—or establish—bodies charged with examining ethical issues and protecting the rights of those affected by AI.
Similarly, the Pope Leo XIV framing stresses safeguards for what guarantees the human person’s balanced growth and asks how to ensure AI truly serves the common good instead of merely concentrating wealth and power.
Drawing together the recurring Catholic principles above, Catholic ethical guidance implies the following development priorities:
A recurrent tension in public debate is whether advanced systems should be treated as exercising agency. Catholic sources acknowledge AI can appear to “choose” or simulate reasoning, but they repeatedly caution against confusing simulation with human moral agency. The practical ethical consequence is that systems should remain tools whose outputs are interpreted and used by accountable human persons according to the common good.
In short, Catholic ethical guidelines for AI development converge on one overarching norm: progress is morally authentic only when it safeguards human dignity and freedom, protects truth and justice, serves the vulnerable and peace, and respects creation through responsible stewardship.