What Makes a Human Human?

AI discussions often become arguments about human uniqueness. A model writes poetry, so humans defend creativity. A system solves difficult problems, so humans defend reasoning. A machine recognizes images, speaks naturally, composes music, or writes code, and each new capability seems to move another boundary.

This is a poor foundation for Christian anthropology. If human dignity depends on outperforming machines, dignity becomes hostage to the next benchmark. It also creates an uncomfortable hierarchy among humans. Many people do not write poetry, solve advanced mathematics, speak several languages, or reason quickly under pressure. Infants, people with profound cognitive disabilities, those living with dementia, and people whose capacities are temporarily impaired do not become less human because their measurable performance differs from someone else’s.

Christian anthropology begins elsewhere: human beings are creatures made in the image of God.

The phrase imago Dei has generated a long history of interpretation. Christians have located its significance in rationality, relationality, vocation, representation, moral capacity, or combinations of these. Responsible theology should not pretend that one disputed interpretation can be lifted from Genesis and turned into a simple anti-AI argument. “Machines cannot do X, therefore X is the image of God” is especially weak. It makes theology chase technical capability.

What the biblical witness does establish with unusual force is the dignity and responsibility of human beings before God. Humans are not valuable because they are the fastest information processors in creation. They are addressed by God, called to faithful creaturely life, placed in relationship, made responsible for action, and destined for an embodied future under God’s redemptive purpose.

Embodiment belongs centrally to this account. Humans do not merely inhabit bodies as disposable containers for cognition. We are born, grow, hunger, sleep, work, suffer, touch, age, and die. We depend on other embodied people before we can speak or reason independently. Our moral lives are shaped through places, histories, families, communities, and vulnerabilities that cannot be reduced to abstract information processing.

This does not mean every valuable interaction must occur face to face. Letters have mediated real Christian friendship for centuries. Telephone calls, video conversations, online communities, and digital teaching can involve genuine relationship. Embodiment is not a slogan for rejecting mediated communication. It is a reminder that human life includes forms of mutual presence, vulnerability, limitation, and shared material existence that cannot simply be declared obsolete because digital interaction becomes convincing.

Creatureliness also means dependence. Modern technological culture often frames dependence as a defect to be overcome. The ideal user can access everything instantly, solve problems alone, automate unpleasant obligations, and minimize reliance on other people. Christian anthropology sees dependence differently. Human beings are dependent on God and interdependent with one another. Need is not always a failure of optimization.

This matters in mission. Learning a language slowly can place a missionary in a posture of dependence on local speakers. Asking a colleague for help can create accountability. Remaining with people through difficulty can form patience and trust. AI can remove unnecessary obstacles in each of these areas, but not every dependency should be treated as a defect.

Moral responsibility is another central feature. Humans answer for what they do with tools. This remains true even when systems become complex enough that no single person can predict every output. Responsibility may be distributed across designers, deployers, leaders, reviewers, and users, but complexity does not make it disappear.

When a church publishes a fabricated quotation generated by AI, “the model made it up” explains the mechanism but not the moral accountability. When a mission organization uploads sensitive information to a third-party system, the provider’s architecture matters, but the organization still bears responsibility for the decision to disclose. When a leader delegates external actions to an agent, automation changes who performs the immediate action without automatically changing who authorized it.

The distinction between agency and responsibility becomes even more important if machines appear increasingly autonomous. Christians do not need to solve the metaphysics of future machine moral agency before holding present human institutions accountable for present deployments. A system may act in ways its user did not specifically script; the user or organization still chose to give it permissions, define objectives, connect tools, or deploy it in a context.

Relational existence also resists reducing humans to isolated cognitive units. We become persons within communities we did not create. Language itself is received socially. Christian identity is ecclesial: believers belong to a body, receive teaching, correction, sacraments or ordinances, care, discipline, and mission within communities. An individualized AI assistant can be extraordinarily capable while still operating within a life whose deepest Christian responsibilities are not private.

This creates an important boundary for synthetic authority. A model may possess more theological information than a particular pastor. It may produce clearer explanations. Neither fact creates pastoral office. Office is not simply a competence score. It involves calling, recognized responsibility, accountability, relationship, and a community capable of correction and discipline.

Human dignity also requires caution when discussing replacement. Economic conversations about AI often ask which workers can be replaced. Ministries can adopt the same logic without noticing. If AI can answer routine questions, why staff a person? If it can generate translations, why train translators? If it can write curriculum, why invest in educators?

Sometimes fewer labor hours really are needed. There is no Christian duty to preserve inefficient processes for their own sake. But Christian organizations should distinguish task replacement from person devaluation. A role may change because tools change; the person remains more than the economic bundle of tasks previously purchased from them.

This is particularly important in Majority World and low-resource contexts. A wealthy organization can treat AI as a way to bypass local expertise: generate translations instead of training translators, generate contextual material instead of listening to local theologians, build software remotely instead of transferring technical capacity. Such workflows may produce short-term output while weakening the human ecosystem that makes ministry sustainable.

A Christian anthropology should instead ask whether technology strengthens the capacity of people to exercise vocation, judgment, and responsibility. The goal is not to preserve every old job description. It is to avoid defining people as obstacles standing between an organization and a cheaper output.

The resurrection gives the account an eschatological dimension. Christian hope is not escape from embodied creatureliness into pure information. The New Testament directs hope toward resurrection, renewed creation, communion with God, and the restoration of persons. Whatever future AI becomes, Christian hope is not fulfilled by merging humanity into a more capable computational system.

This does not settle every philosophical question about machines. It does settle where human worth is located. We do not need to remain cognitively incomparable to our tools in order to remain human. A calculator’s superiority at arithmetic did not diminish human dignity. A chess engine’s superiority at chess did not. A language model’s fluency does not.

The imago Dei is therefore not a shrinking territory around whatever machines still cannot do. It is a theological claim about who human beings are before God.

That claim should make Christians less anxious about technical capability and more serious about human responsibility. The danger is not that a machine becomes competent enough to steal our dignity. The danger is that we begin evaluating ourselves and others by the same performance logic through which we evaluate machines.

Human beings are creatures, embodied persons, morally responsible agents, relational beings, worshipers, neighbors, members of communities, and recipients of grace. Technology can assist these lives. It can also reshape the practices through which they are lived. But no benchmark should become the measure of a human being.¹

Dignity Beyond Performance

The arrival of increasingly capable AI exposes a weakness in many secular and Christian accounts of human uniqueness: they assume the most impressive human cognitive abilities are also the foundation of human worth.

Disability challenges that assumption long before AI does. If rational speed, verbal fluency, independent productivity, or creative output ground dignity, then human beings who possess less of those capacities become difficult to account for. Christian anthropology cannot tolerate that hierarchy.

The image of God belongs to human beings as human beings, not to the top-performing subset. This is why Christians should welcome technologies that expand agency for disabled people without speaking as though technology restores their humanity.

Assistive AI can be profoundly good. Speech generation, image description, personalized interfaces, and cognitive supports can increase participation. Their value lies in removing barriers, not in moving people closer to an imagined norm of worth.

Embodied and Creaturely

Embodiment is sometimes used carelessly in Christian AI critique. “Humans have bodies; AI does not” can become a slogan that does little analytical work.

The stronger question is what embodiment contributes to a specific human good. In pastoral care, embodiment may matter because presence during illness, grief, or crisis includes practical action and shared vulnerability. In language learning, embodiment matters because conversation occurs within gesture, timing, space, facial expression, and social consequence. In church life, bodies gather, eat, sing, baptize, serve, and care.

Digital mediation can participate in many of these goods without reproducing all of them. A video call can sustain a real friendship. It cannot deliver a meal to a sick neighbor. A chatbot can help someone articulate grief. It cannot become the member of the congregation who rearranges a week to sit with them.

This specificity is better than declaring “embodied good, digital bad.” AI encourages a cultural imagination of limitless assistance. The system is available at midnight. It does not tire in the ordinary human sense. It can respond to thousands simultaneously.

Humans cannot. Christian leaders should not allow machine availability to reset expectations for human availability. A pastor does not become negligent because a chatbot can answer at 3 a.m. A missionary does not become inadequate because a model can translate more languages. Human vocation includes limits.

Organizations may even need to protect workers from comparisons with machines. If AI reduces drafting time, leadership should not assume the entire human role should accelerate to machine tempo. Review, relationship, judgment, and rest operate at human scales.

Embodiment matters differently across pastoral care, worship, language, and service; digital mediation can be real without carrying every good of bodily presence. The long-term test is what grows around the tool. The ministry may gain speed while losing understanding, or it may use speed to create margin for better judgment and relationship. Success should therefore be evaluated over time through competence, accountability, resilience, and service rather than through the first impressive demonstration.

Responsibility in a Technological World

Debates about future machine moral status may become important. The current governance problem is simpler.

A ministry can treat a model as if it were morally responsible long before philosophers agree that it is. “The AI chose the wrong person” can become a convenient way to hide the human choices that defined the dataset, permissions, deployment, and review process.

Even if future systems acquire properties that complicate moral philosophy, current institutions remain responsible for the authority they grant.

This is why the book repeatedly asks, “If this goes wrong, who is responsible?”

“The AI did it” is not an adequate organizational answer. Christian organizations should evaluate automation through vocation as well as cost. Suppose an AI can produce first-draft donor communication. That may remove work few people experience as vocationally central. Suppose it can produce sermon manuscripts. The question changes because preaching may be bound up with a pastor’s responsibility to study, interpret, and teach.

The same logic applies to translators, educators, and leaders. A task can be automated without the vocation disappearing. Sometimes automation can strengthen vocation by removing secondary work. Sometimes it can hollow out the practice through which the vocation is exercised.

The relevant question is not whether AI can perform the observable output. It is what human responsibility the output sits inside.

Human Hope Is Not Technical Superiority

Footnotes

1. For theological anthropology that resists performance-based accounts of human worth, see Kilner, Dignity and Destiny: Humanity in the Image of God; Middleton, The Liberating Image: The Imago Dei in Genesis 1; Gay, Modern Technology and the Human Future: A Christian Appraisal ↩