Is Interfaith Consensus Becoming AI’s Moral Operating System?

Oct 17, 2026

Four days ago, faith leaders, technology specialists, policymakers, and academics gathered in Nairobi to discuss the values that should guide artificial intelligence. Less than two weeks earlier, the United Nations Institute for Disarmament Research published a policy brief proposing that convergence among religions could provide a shared ethical foundation for governing AI in military and security settings.

These were separate initiatives. The available evidence does not establish formal coordination between them. Yet their proximity reveals an important movement: religion is no longer being invited merely to comment on artificial intelligence. Religious institutions are increasingly being positioned as partners in constructing its moral legitimacy, governance principles, institutional safeguards, and public trust.

This development must be examined with precision. Artificial intelligence raises real dangers. Human accountability, civilian protection, truthful information, protection of children, and restraint in warfare are morally urgent concerns. Christians should not oppose such goods merely because Muslims, Hindus, Buddhists, Jews, governments, or technology companies also discuss them.

The deeper question is one of authority: Can an ethical framework assembled from contradictory religions provide the moral foundation for technologies that may increasingly influence education, spiritual formation, public information, policing, national security, and war?

Scripture’s answer is clear. People of different religions may cooperate toward limited civic goods, but truth about God, humanity, sin, wisdom, justice, and salvation cannot be constructed through interfaith negotiation. “The fear of the LORD is the beginning of wisdom” (Proverbs 9:10). Wisdom does not begin with consensus. It begins with submission to God.

What Happened in Nairobi?

The October 13 Nairobi meeting was the third regional roundtable of the Faith–AI Covenant, following meetings in New York and Paris. According to the organizer-issued account, participants included representatives of Christian, Muslim, and interfaith networks; the All Africa Conference of Churches; the Evangelical Alliance of Kenya; IBM Research Africa; African AI organizations; universities; and civil-society institutions. Discussions addressed radicalization, algorithmic amplification of grievances, African languages, young people’s moral development, children’s emotional well-being, and Kenya’s national AI strategy (Interfaith Alliance for Safer Communities, 2026).

These concerns are not imaginary. AI systems can produce misinformation, imitate spiritual authority, personalize manipulation, and reproduce the biases of their training material. Systems developed primarily around Western languages and assumptions may also represent African languages, cultures, and religious communities poorly.

Kenya is not approaching this subject in a policy vacuum. Its AI Strategy 2025–2030 includes digital infrastructure, data, research and commercialization, governance, talent, investment, and ethical inclusion (Government of Kenya, 2025). The African Union’s Continental AI Strategy similarly calls for coordinated national approaches, responsible AI development, African participation in global governance, and technologies aligned with African contexts and aspirations (African Union, 2024).

The Nairobi meeting therefore occurred inside a real technological and governmental transformation. Africa is not merely consuming AI. African governments, researchers, churches, businesses, and civil-society organizations are being invited to help define how it should be governed.

Yet the Covenant’s ambitions extend beyond local consultation. Its official roadmap describes a series of roundtables involving New York, Paris, Nairobi, Beijing, Singapore, Bengaluru, Rome, and Abu Dhabi. It says the process will develop shared principles and voluntary pledges for launch in Abu Dhabi in 2027 (Faith–AI Covenant, 2026).

This is not yet a binding global regime. No final covenant has been adopted, and voluntary pledges are not international law. Nevertheless, the process is designed to move from dialogue to principles and from principles to institutional commitments.

From AI Ethics to Military-AI Governance

An additional development makes this moment especially significant. On October 2, the United Nations Institute for Disarmament Research published Faith-Based Perspectives on AI, Security and Ethics. The report grew out of a June 12 dialogue conducted with the Holy See at the Palais des Nations in Geneva. Buddhist, Christian, Hindu, Islamic, Jewish, and Sikh representatives were invited to contribute their traditions’ approaches to AI and security.

The report identifies four areas of convergence: the centrality of the human person, continuing human responsibility, restraint of power, and protection of knowledge and information. It proposes that these areas could contribute to principles, norms, guidelines, corporate responsibilities, and multilateral discussions concerning military AI (Mannsperger & Afina, 2026).

Several recommendations are morally serious. The report argues that responsibility must not be transferred to algorithms, that lethal decisions require meaningful human control, that civilian protection must remain central, and that the chain of responsibility should extend through designers, suppliers, commanders, and policymakers.

Christians can affirm much of this. A machine cannot become the moral scapegoat for a human decision. Adam blamed Eve, and Eve blamed the serpent, but God still held each accountable (Genesis 3:12–19). Scripture declares that “each of us will give an account of himself to God” (Romans 14:12). No commander, engineer, minister, or corporate executive will escape that accounting by saying, “The model decided.”

The World Council of Churches has also confirmed that its central committee moderator participated in the Paris Faith–AI Covenant meeting, where representatives from Anthropic discussed frontier AI and participants considered companionship, vulnerability, authority, social cohesion, and the development of the draft Covenant (World Council of Churches, 2026).

The significant change is therefore not simply that religious people are discussing technology. Religion is being brought into the transition from general ethical discussion to governance frameworks affecting companies, faith institutions, international organizations, national policies, and even military systems.

What Has Changed Since the Earlier Warning?

In May, I argued that AI companies and religious institutions were moving toward a relationship of moral legitimation. The danger was not merely harmful software. It was the emergence of a synthetic moral order that appeared humane while locating authority in pluralist consensus rather than divine revelation (Sangwa, 2026a). Several developments now allow a more precise assessment.

First, the Faith–AI Covenant has progressed from its inaugural New York meeting to consultations in Europe and Africa, with further meetings publicly scheduled across Asia and Rome before the Abu Dhabi launch. Second, the Covenant is no longer framed only around general ethical anxiety. In Nairobi it addressed youth formation, religious representation, radicalization, education, African languages, and national AI policy. Third, UNIDIR has separately moved interfaith ethics into military-AI governance, proposing that religious convergence inform multilateral norms and institutional guardrails. Fourth, faith communities are being treated as sources of public trust, local reach, moral vocabulary, and implementation capacity. They are becoming governance participants rather than outside observers.

What has not been established is equally important. There is no evidence that these initiatives constitute one centralized command structure. There is no adopted global Faith–AI Covenant. There is no proof that the Covenant controls national AI policy. There is no biblical basis for declaring either initiative to be the beast, the False Prophet, the mark of the beast, or the completed religious system of Revelation 17. The responsible category is documented institutional convergence with prophetic relevance, not direct prophetic fulfillment.

Shared Words Do Not Mean Shared Truth

The UNIDIR report acknowledges that words such as dignity, responsibility, consciousness, and personhood carry different meanings across religions. That admission exposes the central theological problem. The Christian grounds human dignity in creation: every human being bears the image of the personal God (Genesis 1:26–27). Humanity’s dignity has been damaged but not erased by the Fall. Its final hope lies not in technological progress or spiritual self-realization, but in redemption through Jesus Christ.

The report also describes Vedantic consciousness as Brahman, the ground of all existence; Hindu accountability through karma and dharma; Buddhist ethical intention; Islamic conceptions of dignity; Sikh service; Jewish deliberation; and Catholic social teaching (Mannsperger & Afina, 2026). These traditions may sometimes reach similar practical conclusions. They do not therefore possess the same doctrine of God, humanity, sin, justice, consciousness, death, or salvation.

A Hindu concept of karma is not biblical judgment. Brahman is not the triune God. Buddhist accounts of craving and consciousness are not the biblical doctrines of sin and the soul. Generic human dignity is not the gospel. Shared vocabulary can support limited civil restraint, but it cannot create shared spiritual truth.

This distinction is essential. Common moral action in creation is possible. Common saving ground in religion is not. Jesus Christ is not one contributor to a global reservoir of spiritual wisdom. He is the eternal Son through whom all things were made and in whom all things hold together (Colossians 1:15–17). He did not say that humanity would reach the Father through a negotiated ethical center. He said, “No one comes to the Father except through me” (John 14:6).

The Biblical Pattern of Babel

Genesis 11 does not condemn cooperation merely because people acted together. Its warning concerns unified human ambition seeking security, identity, and greatness apart from obedient submission to God. The builders possessed common language, technical capability, centralized purpose, and a desire to secure humanity’s future. They said, “Let us make a name for ourselves.” Their civilization was coordinated, innovative, and unified. It was also rebellious.

Modern AI governance is not automatically Babel. International cooperation can restrain violence, improve safety, and protect vulnerable people. But Babel provides the biblical test: Is human capacity being submitted to God, or is humanity attempting to define its own good, secure its own future, and construct its own moral order?

The problem is not that AI has values. Every system reflects values through its data, objectives, prohibitions, rankings, and institutional incentives. The problem is that a pluralist governance process can disguise contested theology as neutral ethics. Once a common moral framework is embedded into models, standards, educational systems, content policies, or military decision processes, theology has not disappeared. It has become less visible.

Public Witness Without Religious Synthesis

Scripture does not require Christians to flee every mixed institution. Daniel served within Babylonian government, yet he refused the king’s food, resisted idolatry, and continued praying when obedience became costly. Paul spoke in Athens, understood pagan ideas, and even cited local poets. But he did not negotiate a shared sacred worldview. He proclaimed the Creator, exposed idolatry, announced coming judgment, and called everyone to repentance because God had raised Jesus Christ from the dead (Acts 17:16–31).

That is the model for Christian engagement with AI governance. Christians may advocate for identifiable human responsibility in warfare. We may defend civilians, children, privacy, conscience, religious liberty, and truthful information. We may contribute technical expertise and expose algorithmic injustice. We may sit at tables with people of every religion. But the Church must not allow its participation to imply that contradictory religions jointly reveal one moral or sacred reality. Witness becomes compromise when Christ’s exclusive authority must be placed outside the room before the conversation can proceed.

The Prophetic Category

Revelation 13 describes a final system in which political authority, deceptive signs, worship, an animated image, and economic exclusion converge. Revelation 17 portrays corrupt religion entangled with political power, wealth, spiritual seduction, and persecution. The current interfaith-AI initiatives do not satisfy those descriptions. They do not presently demand worship, impose the mark, reveal the Antichrist, or prohibit Christians from buying and selling.

Yet they are prophetically relevant because they illustrate an enabling condition: religious institutions, technology companies, governments, and international organizations can now be assembled around common moral language designed to cross national and doctrinal boundaries.

The likely next stage is not an announcement that all religions have officially merged. It is more subtle: common principles, voluntary pledges, institutional training, model-evaluation standards, trusted-partner networks, religious datasets, youth programs, local-language systems, and governance vocabulary that defines which expressions of religion are socially constructive.

These outcomes are responsible inferences from the published roadmaps, not established facts about every future application. The critical question will be whether Christianity is welcomed only as a supplier of compassion, social cohesion, and moral legitimacy while its proclamation of sin, judgment, repentance, and salvation in Christ is treated as divisive. A future religious order need not erase religious institutions. It may preserve them while translating each into a managed ethical function.

What Should the Church Do?

First, Christians should defend genuine human accountability. No lethal decision should disappear inside an opaque model. Designers, suppliers, commanders, governments, and corporate leaders must remain identifiable and answerable.

Second, churches should scrutinize every interfaith AI statement before endorsing it. Ask who defines dignity, personhood, truth, harm, consciousness, and the common good. Ask whether Christian participation requires silence concerning Christ’s unique lordship.

Third, African churches should build strong biblical resources in African languages. If sound Christian teaching is absent from the digital environment, commercial and ideological systems will fill the vacuum. Translation must preserve doctrine rather than flatten it into generic spirituality.

Fourth, pastors and parents must teach young people what AI is and what it is not. It may retrieve information, translate language, or help organize research. It cannot know God, possess a soul, repent, love, worship, receive the Holy Spirit, or exercise spiritual authority.

Fifth, believers working in government, technology, education, security, and law should remain at their posts as faithful witnesses. Withdrawal would leave consequential systems entirely to those who do not acknowledge God. Participation, however, must remain bounded by Scripture.

Finally, the Church must refuse fear. The answer to a synthetic moral order is not panic, hatred, or careless accusation. It is holiness, doctrinal clarity, courageous witness, prayer, evangelism, and readiness for Christ.

The Gospel Cannot Be Reduced to an Ethical Guardrail

Humanity’s deepest problem is not that artificial intelligence lacks a sufficiently inclusive ethical framework. Our deepest problem is sin. All have sinned and fall short of God’s glory (Romans 3:23). No covenant among religions can remove guilt. No governance framework can regenerate the heart. No model can reconcile a sinner to God.

Jesus Christ, the eternal Son of God, became man, lived without sin, died for our sins, was buried, and rose bodily from the dead (1 Corinthians 15:3–4). Salvation is God’s gift, received through repentance and faith in Christ, not through moral consensus or religious cooperation (Ephesians 2:8–9; Acts 4:12).

The world may eventually give AI a shared ethical vocabulary. It cannot give humanity a Savior. As we await the blessed hope and the appearing of Jesus Christ, the Church must remain watchful without sensationalism, engaged without compromise, and peaceful without surrendering truth. Christ may come for His Church at any time. We do not set dates. We do not confuse every conference with prophecy fulfilled. But neither do we sleep while moral authority, religious legitimacy, technological power, and global governance increasingly converge.

Use the tool. Protect the vulnerable. Restrain violence. Demand accountability. Love every neighbor. Test every spirit. And never place Jesus Christ inside a parliament of equal religious voices.

Discussion about this post

Ready for more?