Humans must remain in control of AI, says bishop
Bishop Oscar Cantú of San Jose said humans must retain control over artificial intelligence to protect human dignity and the common good, warning that time to establish oversight is running short. Cantú shared the message during the U.S. bishops’ fall assembly in Baltimore and in an interview about the work of the USCCB’s newly formed AI task force. The task force is working to apply Pope Leo XIV’s vision for human-centered AI, grounded in human dignity and Catholic social teaching. Recent disruptions to U.S. air travel and online services, though not believed to be AI-related, underscored society’s dependence on digital connectivity.
2 hours ago
Bishop Oscar Cantú urged U.S. Catholics to help ensure artificial intelligence remains under human control, warning that its rapid development could affect human dignity, truth, work and warfare. He said the Church’s task force on AI will support education, dialogue with technology companies and advocacy for safeguards. 1
Cantú, bishop of San Jose, California, made the case in an interview and a Nov. 18 address to the U.S. Conference of Catholic Bishops’ fall assembly in Baltimore. He said technology developers expect AI to reach “full speed within six months to a year,” making public education and engagement urgent. 1
The task force was formed after Pope Leo XIV’s encyclical Magnifica Humanitas and is chaired by Cantú. Its members include eight bishops and three lay consultants: theologian and ethicist Charles Camosy, philosopher Meghan Sullivan, and Kim Daniels, who directs Georgetown University’s Initiative on Catholic Social Thought and Public Life. 1
Cantú said AI could encourage a diminished view of human beings by treating people as measurable or marketable in ways that overlook their inherent dignity. He also warned that generated images and disinformation could make it harder for the public to distinguish truth from falsehood, and called on Church institutions to help people verify information and reason together. 1
He highlighted risks for children and young people, including possible effects on critical thinking, friendship, prayer and face-to-face relationships. Cantú called on Catholic educators to develop guidance on prudent AI use for children and youth. 1
The bishop also pointed to concerns about AI’s effects on jobs and the future of meaningful work. In warfare, he said, moral responsibility—particularly in decisions involving autonomous weapons—cannot be handed over to an algorithm. 1
Cantú said major technology companies had shown interest in continued dialogue with the Church and acknowledged the ethical questions surrounding AI. He described this openness as an opportunity for the Church to bring its longstanding teachings about the human person, truth and the common good into public discussion. 1
The task force plans to provide updates and formation opportunities for bishops and diocesan leaders, with webinars expected to begin after the new year. It also aims to share best practices for AI literacy in Catholic education and catechetical programs, and to work with the USCCB’s government relations team on safeguards such as transparency, accountability and human oversight. 1
Cantú concluded by urging the Church to help define what it means to be human in an age of AI, saying that the technology’s development should keep humanity at its center. 1*
Human dignity and the ethical governance of artificial intelligence
The Catholic approach to AI governance begins with a clear principle: AI must remain a human-made tool, governed by people and judged by whether it serves every person’s dignity and the common good. Its speed, usefulness, or ability to perform cognitive tasks does not make it a human person or a moral agent. Ethical governance must therefore protect human responsibility and freedom, prevent exploitation and exclusion, and help technology contribute to people’s integral development.
Catholic teaching grounds human dignity in the person’s existence as created in the image of God—not in intelligence, productivity, independence, or measurable skill. This is important when comparing human intelligence with AI: such comparisons can encourage the mistaken idea that people are valuable in proportion to what they can do. Antiqua et Nova warns against this functionalist view and insists that dignity remains intact even when a person cannot exercise their abilities, including in the cases of an unborn child, an unconscious person, or an elderly person who is suffering.
That principle has practical consequences for AI systems used to assess people, allocate services, or make decisions affecting their opportunities. A system’s score or prediction cannot define a person’s worth. Nor should efficiency become an excuse to treat people as data points whose claims can be ignored. The US bishops accordingly say AI should supplement human work, not replace people or their moral judgments.
AI can generate outputs and influence choices, but the Church’s account distinguishes those functions from human moral agency. People design systems, choose their purposes, deploy them, and decide how to respond to their results. A machine is not a conscience-bearing person capable of freely accepting moral responsibility for its actions.
Between a machine and a human being, only the latter is truly a moral agent—a subject of moral responsibility who exercises freedom in his or her decisions and accepts their consequences.
This is why accountability cannot be evaded by saying that “the algorithm decided.” Responsibility belongs to the people and institutions that shape and use the system. Ethical governance should make it possible to identify who is answerable for an AI-assisted decision and to assess whether the system’s purposes and effects respect people. The US bishops specifically call for human oversight and well-defined accountability, alongside transparency and fair democratic processes.
Catholic evaluation is not limited to whether a system works as intended or is profitable. It asks whether the system’s ends, means, and overall vision respect human dignity and promote the common good. That test applies across the life of an AI system: its design, ownership, operation, regulation, and everyday use.
Pope Leo XIV describes AI as “above all else a tool” and calls for its effects to be assessed in light of the integral development of the human person and society. That development includes not only material welfare but also intellectual and spiritual well-being, protection of each person’s dignity, and respect for the cultural and spiritual diversity of peoples.
The common good also asks who receives AI’s benefits and who bears its costs. The bishops caution that without deliberate ethical and regulatory safeguards, AI may perpetuate marginalization and discrimination, create new forms of poverty, widen the digital divide, or worsen existing inequalities. They argue that AI serves society more fully when poor and vulnerable people can participate in and benefit from its development and use.
A system may increase access to information while still weakening a person’s ability to judge, discern, and choose. Catholic reflection therefore asks not only what AI enables, but how its use affects human freedom and development. Pope Leo XIV has highlighted the technology’s possible effects on critical thinking, learning, interpersonal relationships, and openness to truth and beauty. He also warns that access to large amounts of data should not be confused with the ability to discern meaning and value.
Truth is a governance issue as well as a personal virtue. The bishops point to deepfakes, manipulated news and political information, and the spread of falsehoods as ways AI can undermine both truth and human dignity. Their call for oversight, accountability, and transparency reflects the need to guard against uses that mislead people or distort public decision-making.
The effects on children and young people deserve particular care. Pope Leo XIV stresses that technology should help, not hinder, their intellectual and neurological development and their growth toward maturity and responsibility. This does not mean that every use of AI is harmful; it means governance and education should consider whether young people are being helped to think and decide freely, or encouraged to become passive consumers of machine-generated content.
The responsibility for ethical AI does not belong only to engineers or regulators. Antiqua et Nova says that wise management concerns every level of society and should be guided by subsidiarity—the principle that decisions should involve the appropriate levels of responsibility, rather than being needlessly centralized or detached from those affected. It also emphasizes that human intelligence is needed not only to build technology but to direct its use toward the authentic good of the person.
This makes governance a continuing human task. Developers, owners, operators, regulators, institutions, and users all have roles in ensuring that AI respects dignity and human freedom. Pope Leo XIV calls for serious reflection and responsible governance because AI already affects people’s lives and may concentrate wealth and power if its development is not directed toward the common good.
The Catholic standard, then, is not whether AI appears intelligent or whether it can replace a human task. It is whether people govern it responsibly, preserve human judgment and accountability, protect those most at risk, and direct its power toward a more just and genuinely human society.