After Babel

Why AI may revive the Church’s public voice on human dignity

Artificial intelligence may revive the Church’s public voice not by making society more religious, but by making the question of human dignity harder to avoid.

For most of the digital era, debates about technology have been framed in the language of innovation, productivity, safety, competition, regulation and risk. These are necessary vocabularies. They help policymakers ask whether systems are accurate, lawful, fair, explainable, secure or accountable.

But artificial intelligence is now pushing public debate into deeper territory. It is forcing societies to ask not only what machines can do, but what human beings are. What is the difference between intelligence and wisdom, communication and relationship, simulated care and genuine care, prediction and judgment, influence and manipulation, and autonomy and behavioural optimisation?

These are not exclusively religious questions but are civic, legal, philosophical and democratic questions. But they are also questions on which religious traditions, including the Catholic tradition, have long developed a public discourse.

That is why Pope Leo XIV’s recent encyclical, Magnifica Humanitas: On Safeguarding the Human Person in the Time of Artificial Intelligence, is significant beyond the internal life of the Catholic Church. Its relevance lies not in providing technical answers to AI governance, but in articulating a broader concern that technological progress must be assessed by reference to the dignity, freedom and flourishing of the human person.

The encyclical places AI within the tradition of Catholic Social Doctrine and explicitly connects the present technological moment with earlier social transformations, including the industrial upheavals addressed by Rerum Novarum. Its central claim is that AI should be evaluated not merely as a tool of efficiency, but as a force capable of reshaping power, work, truth, education, social relations and democratic life.

For a secular digital ethics organisation, the importance of this intervention is not theological authority. It is the fact that a major global institution is placing human dignity, vulnerability and the common good at the centre of AI ethics at a time when many public debates remain dominated by speed, scale and market competition.

AI and the return of the human question

AI has made old questions newly practical.

Until recently, the claim that human judgment, care, creativity or conversation were distinctive features of human life could be treated as largely philosophical. Today, those claims are becoming regulatory and social questions.

AI systems can now produce fluent language, simulate empathy, generate intimate responses, assist in professional decision-making, shape information environments, classify people, recommend choices, and increasingly mediate access to services and opportunities. They do not need to be conscious in order to influence consciousness. They do not need to understand the human person in order to profile, predict or persuade one.

This is why dignity matters. Not as a slogan, but as a limiting principle.

A dignity-based approach asks whether systems treat people as persons or merely as data subjects, consumers, users, workers, patients, students, voters, risk scores or behavioural targets. It asks whether a system preserves agency or weakens it. It asks whether it supports human judgment or displaces it. It asks whether it exploits dependence, loneliness, confusion or vulnerability. It asks whether people can understand, contest and resist the decisions being made about them.

The encyclical’s contribution is to insist that AI cannot be assessed only by what it produces. It must also be assessed by what it presupposes about the human person.

The image of Babel

The title of this piece draws from one of the encyclical’s central images: Babel.

In the document, Babel functions as a warning about technological ambition detached from humility, plurality, responsibility and the common good. The opposing image is the rebuilding of Jerusalem: a model of shared responsibility, reconstruction and social cooperation. The contrast is not between technology and anti-technology. It is between two ways of building something. One based on domination, homogenisation and power and the other based on human dignity, participation and the common good.

As a secular metaphor, Babel is useful because it captures one of the central risks of the AI age that the construction of systems so large, fast and powerful that human beings become adapted to the system, rather than the system being ordered toward human life.

This risk is visible across several domains.

In the workplace, AI may support workers, reduce drudgery and increase safety. But it may also intensify surveillance, deskill labour, displace roles and subject human beings to machine-paced productivity metrics.

In education, AI may support access, tutoring and learning. But it may also weaken attention, reduce the effort required for inquiry, and blur the distinction between acquiring information and developing judgment.

In public discourse, AI may support translation, accessibility and communication. But it may also accelerate disinformation, synthetic media, manipulation and the collapse of shared factual reality.

In personal life, AI may offer assistance and companionship. But it may also simulate care in ways that deepen dependency, loneliness or emotional confusion.

In governance, AI may improve public administration. But it may also create opaque systems through which individuals are classified, excluded or disadvantaged without meaningful explanation or appeal.

These concerns are not speculative. They are already present in debates on platform governance, data protection, labour rights, online safety, AI regulation, children’s digital rights, democratic resilience and automated decision-making.

Why the Church’s voice may become newly audible

To say that AI may revive the Church’s public voice is not to suggest a return to religious dominance. Nor is it to suggest that the Church has unique ownership of dignity, ethics or truth in public life.

The point is more limited and more interesting.

AI is exposing a gap in contemporary public language. Many institutions can speak about innovation. Many can speak about risk. Many can speak about compliance. Fewer can speak persuasively about what kind of human life technology should serve.

The Catholic tradition, whatever one’s religious position, has a developed body of social thought on dignity, solidarity, subsidiarity, the common good, labour, human rights, poverty, peace and the moral limits of economic power. These concepts are not a substitute for law or technical governance. But they can enrich public debate by asking questions that market-led and state-led frameworks sometimes underplay.

That may be where the Church’s public relevance returns. Not as a governing authority, but as one voice among others pressing the human question.

In secular terms, the Church may become more audible because AI has made the anthropology of technology unavoidable.

The limits of purely technical ethics

Much of AI governance is necessarily technical. Systems require testing, documentation, risk management, cybersecurity, audit, transparency, procurement standards, liability rules and enforcement mechanisms.

But technical governance alone cannot answer the deepest questions AI raises.

A system may be transparent and still manipulative.

It may be accurate and still unjust.

It may be efficient and still dehumanising.

It may be personalised and still corrosive of autonomy.

It may comply with formal consent rules while operating in an environment designed to capture attention and weaken self-direction.

This is why public AI ethics needs more than procedural safeguards. It needs substantive commitments about the human person.

The encyclical contributes to this wider debate by challenging the assumption that technological progress is automatically human progress. It warns that technology is never neutral in practice, because it reflects the priorities of those who design, finance, regulate and deploy it.

That claim should be recognisable even in fully secular terms. AI systems embody choices: what to optimise, what to measure, what to ignore, what to classify, what to predict, what to recommend and what to make visible or invisible.

Every one of those choices has moral and political significance.

Power, not just intelligence

One of the strongest aspects of Magnifica Humanitas is that it treats AI not merely as artificial intelligence, but as a form of power.

The encyclical highlights the concentration of technological power in the hands of major private and transnational actors, including those who control platforms, infrastructure, data and computational capacity. It warns that these actors may shape access, visibility, communication and economic opportunity in ways that exceed the capacity of many public institutions to oversee.

This is a core issue for secular AI governance. The most important question is not whether machines will become “like humans.” The more immediate question is whether human beings will be governed by systems they cannot understand, influence or contest.

A dignity-centred approach therefore requires attention to:

• concentrated platform power
• opaque algorithmic decision-making
• behavioural manipulation
• synthetic media and disinformation
• workplace surveillance and automation
• children’s exposure to addictive or exploitative design
• data extraction and asymmetrical profiling
• automated exclusion from services or opportunities
• AI systems used in policing, borders, welfare, credit, employment and warfare

These are legal, democratic and institutional questions. They are also dignity questions.

A secular reading of human dignity

For CDE, the most useful way to engage with the encyclical is not as a religious text to be endorsed, but as a public intervention into the dignity debate.

Human dignity can be understood in secular terms as the claim that every person has a worth that precedes usefulness, productivity, intelligence, social status or economic value. It means that people must not be reduced to instruments of profit, efficiency, control or experimentation. It means that systems affecting human lives must be designed and governed around agency, fairness, vulnerability, accountability and respect.

This secular understanding overlaps with, but does not depend upon, theological claims. That overlap is important. In pluralist societies, religious and secular actors may disagree about ultimate foundations while still converging on important protections for human dignity.

AI ethics needs that kind of convergence.

The danger is not that one tradition will have all the answers. The danger is that no institution will be willing to insist on the question strongly enough.

What should we take from this moment

The significance of Magnifica Humanitas is that it places AI within a broader human question.

It refuses to treat AI ethics as merely a matter of system performance. It connects AI to truth, work, education, social justice, children, human rights, peace, dependency, exploitation, democratic life and the common good. Whether one accepts its theological foundations or not, this is the kind of integrated analysis that AI governance urgently needs.

For CDE, the key lesson is that dignity must become operational.

It is not enough to invoke “human-centred AI” in principle. We need to ask what that means in design, deployment, procurement, audit, enforcement and remedy.

A dignity-based AI framework should ask:

• Does this system strengthen or weaken human autonomy?
• Does it preserve meaningful human judgment?
• Does it exploit vulnerability, dependency or emotional need?
• Does it simulate care where real care is required?
• Does it affect children or other vulnerable groups differently?
• Does it allow explanation, contestation and redress?
• Does it concentrate power in ways that undermine democratic participation?
• Does it treat workers and affected communities as stakeholders or as inputs?
• Does it support truth, trust and public reason?
• Does it make human life more humane?

These questions are not religious. They are foundational to any serious democratic approach to AI.

After Babel

The age of AI is a construction site. We are building systems that may reshape knowledge, work, intimacy, education, public discourse, law, warfare and democracy.

The question is not whether we build. We already are building.

The question is whether we build systems that reduce the person to data, behaviour and output, or systems that remain answerable to human dignity.

That is why the Church’s voice may become newly relevant in public debate. Not because AI ethics should become religious. Not because religious institutions should govern technology. But because the AI age has made it clear that societies need more than technical expertise, market incentives and legal compliance. They need a public moral language capable of defending the human person against reduction.

In that sense, the Church’s intervention should be read as part of a wider civic conversation. It is one voice among many asking the question that should stand at the centre of AI governance:

What kind of human future are we building, and who gets to decide?