WHO WILL DISCIPLE BLACK PEOPLE IN THE AGE OF ARTIFICIAL INTELLIGENCE?
The Black Church, Algorithmic Formation™, and the Struggle for Black Moral Imagination
Justice Scholars Society | A DMoPA Research Essay
Abstract
Artificial intelligence increasingly mediates the environments in which individuals encounter knowledge, exercise judgment, negotiate identity, and experience institutional power. For Black ecclesial institutions, this development poses a challenge that extends beyond technological adoption and algorithmic bias to the formation of human dignity, moral responsibility, and communal agency. Employing the Douglassian Methodology of Prophetic Agitation (DMoPA), this essay develops a constructive Black public-theological framework that exposes contradictions between institutional professions and practices and directs moral critique toward structural reconstruction. Integrating Black liberation theology, Womanist ethics, disability theology, and interdisciplinary AI scholarship, the analysis advances three interrelated constructs: Algorithmic Formation™, which examines computational participation in formative environments; the Non-Computability of Human Worth™, which rejects computational classification and performance as determinants of intrinsic dignity; and Performance–Formation Fragmentation™, which identifies the unvalidated theoretical risk that technologically enhanced performance may become disconnected from human competence and accountability. Extending beyond ecclesial practice, the essay interrogates labor, data, surveillance, cultural extraction, and technological infrastructure as interconnected sites of institutional power. It proposes Protection + Participation + Ownership as a reconstructive framework for strengthening Black technological agency while locating human flourishing beyond technological efficiency and economic accumulation. The essay concludes that Black ecclesial institutions should cultivate interdisciplinary partnerships that strengthen moral discernment, institutional accountability, and communal self-determination without surrendering theological responsibility to computational systems.
Keywords: Artificial Intelligence; Black Church; Black Public Theology; Womanist Ethics; Algorithmic Formation; Human Dignity; DMoPA; Technological Agency.
INTRODUCTION — FORMATION IS THE QUESTION
Religious communities often encounter artificial intelligence as a question of adoption: Should pastors use generative AI? Should congregations automate administrative tasks? Can AI assist with biblical research, education, accessibility, communication, or pastoral care? These questions matter, but they come too late. They assume AI becomes theologically significant primarily when the Church decides to use it.
AI already mediates information, education, work, communication, culture, and institutional decisions. Generative systems also synthesize, interpret, and produce content. Pew Research Center found that 49% of Black adults surveyed in February 2026 reported using AI chatbots, including 24% who used them daily (Gottfried et al., 2026). These findings document reported use, not the magnitude, direction, or durability of effects on religious formation. Technological presence is not formative causation.
This essay argues that Black ecclesial institutions should understand artificial intelligence not merely as a technology to adopt or a risk to regulate, but as part of an emerging environment of human formation and political economy. Drawing on Black theological anthropology, Womanist ethics, Black ecclesiology, and contemporary AI scholarship, it develops a public-theological response centered on human dignity, accountable technological augmentation, and Black technological agency. Its reconstructive framework is organized around Protection + Participation + Ownership.
Formation is the essay's organizing category. “The Black Church” denotes a historically recognizable yet internally heterogeneous constellation of Black Christian traditions and institutions, not a monolithic ecclesial subject. This argument neither assumes that all Black Americans are Christian nor claims universal ecclesial authority. Black churches differ across denominations, gender, sexuality, class, geography, generation, politics, worship traditions, and technological capacity. They have also reproduced exclusions and hierarchies that require criticism from within Black religious traditions.
Methodologically, the essay employs the Douglassian Methodology of Prophetic Agitation (DMoPA) as a form of constructive public theology. DMoPA reconstructs a pattern evident in Frederick Douglass's public work: exposing contradictions between moral professions and institutional practices, grounding indictment in evidence, agitating normalized arrangements of domination, and pressing criticism toward structural reconstruction. Douglass's Fourth of July address exposed the contradiction between America's profession of liberty and slavery; his Reconstruction-era writings pressed for political rights and institutional repair (Douglass, 1852, 1866). DMoPA functions here as an interpretive and constructive methodology, not an empirical research design.
Applied to AI, the method asks where promises of democratization, efficiency, intelligence, and progress run up against exclusion, extraction, dependency, diminished accountability, or unequal burdens. It also operates reflexively: churches cannot credibly interrogate technological domination while refusing scrutiny of their own authority. Who decides? Who benefits? Whose voice is absent? Who can contest? Who remains accountable?
The argument proceeds through formation, dignity, accountable augmentation, political economy, and agency. Its constructs are theoretical propositions inviting investigation rather than validated causal models. The Black Church need not compete with AI for informational superiority. Its vocation is to help form people capable of deciding what intelligence is for, what power owes to persons, and what kind of technological future deserves to be built.
I. THE BLACK CHURCH AS AN INSTITUTION OF INTERPRETATION
Black Ecclesial Formation Within the Algorithmic Age
Before artificial intelligence could generate sermons, synthesize theological arguments, or answer biblical questions, Black Christian communities had already spent generations grappling with a fundamental interpretive problem: How does a people understand itself faithfully within institutions that have repeatedly misinterpreted its humanity?
That question explains why Black ecclesial history matters to AI.
Lincoln and Mamiya's (1990) landmark study situates historically Black churches within relationships that extend beyond formal worship into politics, economics, gender, youth, music, and broader Black culture. Across diverse periods and traditions, Black churches have engaged in education, mutual aid, leadership development, political organization, cultural preservation, social protest, and communal interpretation.
Their significance does not lie in institutional innocence. It lies partly in creating spaces where Black communities could contest dominant interpretations of God, Scripture, humanity, citizenship, suffering, freedom, and social order. When American Christianity accommodated racial domination, Black religious thought exposed the contradictions between Christian profession and racial practice. Preaching, testimony, sacred music, biblical interpretation, and communal memory offered alternative accounts of dignity and possibility.
The Black Church has therefore functioned, among other things, as an institution of interpretation.
Digital mediation is not new to this history. Gault (2022) shows that Black Christians have already formed religious identities, relationships, communities, and practices in digital spaces, both within and beyond physically located churches. Artificial intelligence should therefore be understood not as Black Christianity's first encounter with technological mediation but as a significant development within an existing digital religious ecology.
Generative AI changes the character of mediation. Computational systems no longer merely direct users to information. They can rank, summarize, personalize, synthesize, generate, and interpret it conversationally. A user may receive a coherent theological, historical, or social explanation without clearly seeing which interpretive decisions produced the synthesis. Sources can disappear into fluent prose; disagreement can be compressed; contested interpretations can appear settled.
The issue is therefore larger than hallucination. It concerns formation through mediation.
Within DMoPA, Algorithmic Formation™ describes the cumulative participation of computational systems in shaping human attention, perception, judgment, knowledge, desire, and social imagination through processes such as selection, ranking, recommendation, personalization, generation, summarization, and interpretation.
The concept does not claim that algorithms manufacture Black consciousness. Families, churches, schools, economies, culture, history, and human agency remain formative. AI enters that ecology without replacing it.
Its possibilities can also be emancipatory. AI may expand access to scholarship, tutoring, translation, accessibility, research, entrepreneurship, and institutional capacity. That possibility is precisely why technological fear is inadequate.
Butler (2024) argues that Black theology and Womanism are already engaged with artificial intelligence, including questions of white supremacy, surveillance, policing, consciousness, and God. DMoPA therefore does not inaugurate Black theological engagement with AI. Its narrower intervention is to place formation at the center of Black ecclesial engagement with intelligent technologies.
That changes the governing questions.
Not simply whether an AI system contains racial bias, but what forms of judgment are necessary when computational systems repeatedly mediate knowledge.
Not simply whether Black communities possess technological access, but whether access develops agency or sophisticated dependency.
Not simply whether AI can generate theological information, but who verifies its claims, identifies its assumptions, recognizes its absences, and remains accountable for what is believed or done.
Womanist ethics intensifies these questions by interrogating apparently universal categories. Terms such as the user, human preference, helpfulness, safety, and human values can appear neutral while concealing the social locations from which their definitions emerge. The appropriate response is neither predetermined suspicion nor technological innocence. It is disciplined interrogation.
Here DMoPA's Prophetic Verification becomes indispensable. Moral urgency cannot become permission for evidentiary carelessness. Black communities' historical experience of misrepresentation should make Black ecclesial institutions more—not less—committed to identifying sources, testing claims, distinguishing evidence from interpretation, and revising conclusions when evidence requires it.
The Black Church's technological vocation therefore cannot depend upon possessing more information than artificial intelligence.
Its distinctive responsibility is interpretive:
What deserves trust?
Which authorities require scrutiny?
What should never be delegated?
When does efficiency conflict with dignity?
What responsibilities accompany technological power?
These questions lead toward anthropology: human significance cannot rest on identifying activities machines cannot perform.
It requires a stronger account of the human person.
II. THE NON-COMPUTABILITY OF HUMAN WORTH™
Black Theological Anthropology After Machine Intelligence
If formation is the organizing problem of this essay, theological anthropology supplies its normative center. Artificial intelligence confronts Christian theology with a question more fundamental than whether machines can think, create, reason, or converse:
What makes a human being worthy when intelligence itself can increasingly be technologically reproduced, supplemented, and exceeded in particular domains?
Defending human uniqueness through capacities machines lack is unstable: every advance moves the boundary of supposedly exclusive human competence.
Dignity Before Performance
Black people have repeatedly confronted social orders that converted socially constructed or purportedly measurable differences into hierarchies of human value. Enslavement assigned economic value to Black bodies while denying enslaved persons freedom over themselves. Scientific racism attempted to naturalize racial hierarchy through measurements of bodies, intelligence, heredity, and alleged civilizational capacity. Segregation translated racial classification into unequal citizenship, education, mobility, residence, and institutional power.
The deeper theological failure was assuming that measurement could establish worth.
James Cone's (1970, 1975) liberation theology locates theological reflection within Black historical experience and the struggle against oppression. Cone did not formulate AI ethics; DMoPA extends his concern with liberation to technological structures of power.
Technical performance cannot establish theological legitimacy.
Womanist Ethics and the Universal Human
Katie Cannon (1988) develops Black women's moral wisdom and agency in critical relation to dominant ethical traditions. Delores Williams (1993) examines survival, surrogacy, and imposed suffering through her Womanist reading of Hagar. Neither formulated contemporary AI ethics. DMoPA extends their distinct concerns by asking whose experiences disappear behind apparently universal categories such as the user, helpfulness, and safety, and whose labor or burdens remain invisible in accounts of technological efficiency.
Womanist ethics therefore contributes more than demographic inclusion:
Universal technological claims must be interrogated from the social locations most likely to disappear inside their universality.
Universality cannot establish its own innocence.
The Non-Computability of Human Worth™
Against performance-based anthropology, DMoPA proposes the Non-Computability of Human Worth™:
No computational description, prediction, score, classification, or comparison possesses sufficient moral authority to exhaust or determine the worth of a human person.
Non-Computability is used normatively, not as a technical proposition in theoretical computer science. Human behavior can be modeled. Cognitive capacities can be measured. Statistical systems can make useful predictions. Artificial intelligence can outperform human beings in particular tasks.
The distinction is between description and valuation.
A credit model may estimate repayment probability without determining dignity.
An employment system may predict performance without exhausting vocational possibility.
A risk model may classify someone statistically without becoming the total meaning of that person.
A generative model may produce prose more fluently than a particular student, scholar, or preacher without acquiring greater moral worth.
The question is what measurement is authorized to mean.
This yields the manuscript's anthropological hinge:
Human dignity cannot depend upon outperforming machines because human dignity never legitimately depended upon outperforming other human beings.
Dignity must not depend upon superiority.
Disability Theology and the Critique of Capacity
Disability theologians challenge anthropologies that make normative capacities conditions of full personhood. Eiesland (1994) critiques exclusionary assumptions about embodiment through a liberatory account of the disabled Christ; Yong (2007) examines disability through creation, imago Dei, and ecclesial belonging; Reynolds (2008) emphasizes vulnerability, dependence, and hospitality. Their arguments differ, but together they expose the inadequacy of equating dignity with intelligence, productivity, or independence.
Artificial intelligence therefore does not create the problem of performance-based humanity.
Machine intelligence makes the inadequacy of that anthropology newly visible.
The task is to refuse hierarchies in which capacity determines dignity.
Embodiment, Relationship, and Moral Ends
Black theological anthropology also requires attention to embodiment. Black bodies have historically been sites upon which political, economic, medical, cultural, and religious power has been exercised. Any account of personhood that treats embodiment as incidental to intelligence risks returning to an abstraction Black experience has repeatedly shown to be inadequate.
Embodiment does not guarantee wisdom: pastors and institutions can abuse power.
The stronger claim is that embodiment creates forms of situatedness, vulnerability, relational consequence, and accountability relevant to theological judgments about relationship.
Likewise, personalization cannot constitute an adequate anthropology. Computational systems may learn what an individual prefers, but human flourishing cannot be reduced to preference satisfaction. Black religious and communal traditions locate persons within relationships of kinship, congregation, neighborhood, history, memory, and obligation.
The theological question is therefore:
What kind of “me” is technology helping to form in relation to “us”?
Community can interrupt preference. Covenant creates obligation. Justice may require relinquishing advantage. Solidarity may demand attention to people whose needs do not optimize one's individual experience.
This exposes the distinction between intelligence and wisdom.
AI can increasingly optimize means. It cannot establish the moral legitimacy of an end merely by optimizing it efficiently. The least expensive option may not be the most humane. The most persuasive communication may be manipulative. The most profitable decision may exploit vulnerability.
Contemporary Christian AI scholarship similarly resists reducing human flourishing to enhanced intelligence alone, emphasizing dignity, embodiment, relationality, and moral life. DMoPA's contribution is to place that anthropological concern explicitly within Black histories of racial classification, labor valuation, and denied humanity.
The resulting principle is:
Intelligence can optimize means; moral responsibility remains accountable for interrogating the ends.
Work Without Ontological Depreciation
AI may augment workers, automate tasks, transform occupations, and create new forms of employment. The distribution remains uncertain. Yet one theological principle does not depend upon predicting labor markets correctly:
Economic displacement is not ontological depreciation.
A worker whose task loses market value has experienced an economic change.
The worker has not become less human.
This carries particular resonance within Black history, where economic systems repeatedly assigned value to Black labor while withholding equal recognition of Black personhood. Economic usefulness and human dignity have never been interchangeable categories.
Black ecclesial institutions therefore face two responsibilities simultaneously: preparing communities for technological labor transitions while refusing a social order that interprets declining market demand as declining human significance.
Classification Must Not Become Destiny
Artificial intelligence increasingly shapes classification systems in credit, fraud detection, employment, education, health, and institutional risk. Classification itself is not inherently unjust.
The danger arises when classification acquires excessive authority over human possibility.
The relevant questions extend beyond statistical accuracy:
Who can challenge the classification?
Can underlying information be corrected?
What opportunity disappears because of it?
Who audits the system?
Does a probabilistic judgment become an institutional certainty?
Can the person escape the category?
Does prediction become destiny?
The Non-Computability of Human Worth does not prohibit measurement.
It establishes a boundary:
A probability may describe something about a human being; it must never become the total meaning of the human being.
The Anthropological Boundary
The Black Church need not prove human superiority to machines.
Black theological traditions bring historical force to another claim because Black communities have repeatedly confronted systems that translated alleged differences in intelligence, productivity, race, gender, bodily capacity, legal status, and economic usefulness into hierarchies of dignity.
Womanist ethics asks whose experience disappears inside universality.
Disability theology challenges normative capacity as a condition of personhood.
Black historical experience demonstrates how classifications can acquire institutional power over human possibility.
The resulting boundary is moral rather than technological:
A human being is always more than the measure assigned to them.
Human worth was never contingent upon winning that competition.
That theological anthropology establishes what must remain protected when artificial intelligence enters ecclesial practice.
III. WHEN INTELLIGENCE ENTERS THE SANCTUARY
Accountable Augmentation, Preacher Formation, and Ecclesial Authority
If human worth cannot be reduced to computational performance, a related ecclesiological question follows: What happens when computational systems begin performing tasks historically associated with ministry?
Artificial intelligence can assist biblical research, compare translations, summarize scholarship, generate educational materials, transcribe sermons, support accessibility, organize information, and produce language resembling sermons, prayers, devotionals, and pastoral responses. These capabilities raise questions not merely about usefulness but about authority, formation, responsibility, and the meaning of ministerial practice.
The governing distinction is:
AI in ministry is not the same thing as AI as ministry.
Artificial intelligence may assist ministerial practices without thereby inheriting pastoral vocation, covenantal responsibility, spiritual authority, or ecclesial accountability. Technical capability does not itself confer ministerial authority.
This is not an argument for clerical exceptionalism. Human ministers can preach falsely, manipulate congregations, mishandle confidential information, abuse authority, and provide destructive counsel.
The distinction is accountable authority, not human good versus machine bad.
Preaching: Performance and Formation
Black homiletical scholarship, including Mitchell (1990), LaRue (2000), and Thomas (2013), resists reducing preaching to the transmission of religious information. Black preaching emerges through relationships among Scripture, preacher, congregation, Black experience, theological imagination, culture, communal memory, embodiment, and proclamation.
Generative AI can now participate in sermon preparation through research assistance, translation comparison, historical synthesis, organization, editing, argument testing, and rhetorical development. For bivocational pastors and resource-constrained ministries, such capabilities may materially expand access to scholarly resources.
The critical distinction is between the sermon as artifact and the preacher as person undergoing formation.
Within DMoPA, Performance–Formation Fragmentation™ names:
the risk that technological enhancement of observable performance may become decoupled from the development or retention of human competencies necessary to understand, evaluate, reproduce, or take responsibility for that performance.
This is a theoretical risk, not an established effect.
Research on cognitive offloading and automation bias demonstrates that external tools can redistribute cognitive effort and influence monitoring and judgment without necessarily producing deskilling (Parasuraman & Manzey, 2010; Risko & Gilbert, 2016). Offloading is not inherently destructive. AI may strengthen formation through unfamiliar scholarship, argument testing, and deeper study.
A preacher may not need to memorize every historical fact. The preacher does need sufficient historical and interpretive judgment to recognize false information.
A minister need not search every database manually. The minister must remain capable of distinguishing credible scholarship from fabricated attribution.
AI may sharpen prose. Rhetorical sophistication cannot compensate for theological misunderstanding.
The governing questions are:
Can the preacher verify the sources?
Understand and defend the interpretation?
Identify credible alternatives?
Recognize theological assumptions?
Remain responsible for what is proclaimed?
The standard is accountable augmentation.
Technological assistance may extend human capacity; it must not obscure human responsibility.
Pastoral Care: Assistance Is Not Covenant
AI systems can generate language users experience as attentive, patient, and emotionally supportive. Recent pastoral-psychology scholarship explores the potential and limitations of AI-mediated assistance without establishing its equivalence to covenantal care (Sande, 2026; Tretter & Glatzel, 2026). AI may help people articulate concerns, locate resources, prepare questions for a pastor or clinician, or receive conversational support when human assistance is unavailable.
These benefits do not establish equivalence with a covenantal pastoral relationship.
Pastoral care can involve durable responsibility, confidentiality, communal recognition, knowledge of family and congregational history, embodied accompaniment, and accountability for what follows a conversation.
Therefore:
Relational simulation must not be casually equated with relational covenant.
Embodiment matters without guaranteeing virtue. An embodied pastor can be negligent or abusive. The stronger claim is that embodiment creates forms of situatedness, vulnerability, relational consequence, and accountability that may remain ecclesiologically significant.
AI can augment pastoral care.
Its availability should not excuse pastoral disappearance.
Theological Fluency Is Not Theological Neutrality
A congregant can ask a generative system what a biblical passage means and receive a coherent response within seconds. The response may synthesize historical scholarship, denominational traditions, theological assumptions, and popular interpretations.
Fluency can create an appearance of neutrality.
Theological fluency must not be mistaken for theological neutrality.
Baptist, Pentecostal, Catholic, Reformed, Womanist, Black liberationist, evangelical, historical-critical, and other traditions do not approach every theological question the same way. Generative synthesis may obscure those differences precisely because synthesis sounds comprehensive.
This requires theological verification:
What sources support the interpretation?
Can they be located?
Does the scholar actually make the attributed claim?
Which traditions are represented?
Which are absent?
What assumptions structure the answer?
What remains contested or uncertain?
AI may democratize access to biblical scholarship and theological traditions previously mediated largely through clergy, universities, seminaries, or specialized libraries. That can challenge poor preaching and unhealthy clerical gatekeeping.
The appropriate response is not restoration of informational monopoly.
It is distributed theological competence.
Congregations should become better able to test both preacher and machine.
From Digital Religion to Generative Religion
Digital-religion scholarship—and Erika Gault's work on Black digital religion specifically—demonstrates that Black Christians have already formed identities, relationships, communities, and religious practices through digital spaces within and beyond physically located churches.
Generative AI changes this mediation rather than inaugurating it.
Digital platforms largely enabled people and institutions to produce, distribute, discover, and interact around religious content. Generative systems increasingly participate in producing, synthesizing, personalizing, and interpreting the content itself.
That development intensifies the questions:
Who—or what—is speaking?
From which sources?
Through which assumptions?
Under whose responsibility?
Congregational Data as Sacred Trust
An immediate ecclesial AI concern is data.
Churches possess unusually sensitive information: giving records, prayer requests, pastoral-care information, health concerns, family crises, children's information, addresses, attendance histories, and religious commitments.
AI-enabled platforms can make such information easier to organize and analyze. They can also expand how information is processed, retained, inferred from, transferred, or repurposed.
Henderson’s (2026) conceptual study of intelligent environments in Black Baptist churches raises concerns involving privacy, trust, accessibility, participatory governance, data minimization, and surveillance; it does not establish empirical effectiveness of the proposed systems.
DMoPA's theological extension is:
How a church governs information entrusted by congregants is part of how the church exercises power over persons entrusted to its care.
Data governance is therefore not merely an IT responsibility.
It is ecclesial ethics.
Churches should know where information is stored, who can access it, whether vendors retain it, whether it may be used for model training, how long it persists, and what information should never enter external generative systems.
Convenience does not eliminate confidentiality.
Black Sacred Memory: Preservation Without Extraction
AI also offers extraordinary possibilities for Black sacred archives.
Sermon recordings, Church minutes, anniversary books, funeral programs, photographs, choir recordings, missionary records, civil-rights correspondence, oral histories, and pastoral manuscripts can potentially be transcribed, restored, indexed, translated, and made searchable.
Preservation nevertheless raises governance questions.
Who controls the digitized archive?
Who may train models on it?
Who may reproduce the voice or likeness of a deceased preacher?
Who determines whether Black sacred music or sermonic style becomes generative material?
Community archival scholarship emphasizes representation and stewardship (Caswell et al., 2017), while U.S. Copyright Office reports distinguish copyrightability, training, and digital-replica issues (U.S. Copyright Office, 2024, 2025). These sources do not establish universal legal prohibitions on AI training.
The theological claim is narrower:
Technological capacity does not create moral entitlement to sacred or cultural inheritance.
And:
Digitization without governance can turn preservation into extraction.
Those propositions bridge ecclesiology and political economy.
The Church Must Submit to Its Own Test
Black churches cannot credibly interrogate technological power while treating ecclesial power as inherently trustworthy.
The same questions directed toward AI systems must be directed inward:
Who decides?
Who benefits?
Whose voice is absent?
Who can challenge authority?
Who remains accountable?
These questions include gender, sexuality, disability, class, generation, clergy power, and institutional hierarchy. A Church cannot condemn algorithmic exclusion while refusing to examine its own exclusions.
DMoPA must operate reflexively:
A Black ecclesial critique of technological domination loses credibility if Black churches refuse equivalent scrutiny of their own structures of authority.
That is a condition of credible Black ecclesial witness.
The Ecclesial Boundary
AI may strengthen ministry, access, research, and archival preservation.
None of those possibilities requires surrendering ecclesial distinctions.
Capability is not vocation.
Fluency is not neutrality.
Simulation is not automatically covenant.
Information is not formation.
Augmentation does not eliminate accountability.
The governing principle is therefore:
Technological systems may extend human capacity, but identifiable human and institutional responsibility must remain proportionate to the consequences of their use.
That principle exposes the inadequacy of examining artificial intelligence only within the sanctuary. The technologies assisting churches exist within larger systems of labor, data, capital, infrastructure, ownership, and environmental consequences.
The analysis must follow them outward.
IV. FROM THE PROMPT TO THE POWER PLANT
Labor, Data, Infrastructure, and the Political Economy of Artificial Intelligence
If artificial intelligence raises questions about formation within the sanctuary, its theological significance cannot end at the interface. A user enters a prompt and receives a response, but that exchange depends on labor, data, intellectual property, capital, semiconductor production, electricity, water, land, data centers, and regulatory authority. A Black public theology must ask how technological power is produced, distributed, and governed. Who supplies the labor and data? Who owns the productive assets? Who bears the material costs? Who participates in decisions and can contest the consequences? A serious Black theology of AI must follow technology from the prompt to the power plant.
Labor Transition: Who Can Move When Work Moves?
Debate about AI and employment often oscillates between technological abundance and mass displacement. The more defensible conclusion is that effects vary by task, occupation, industry, and institutional conditions (Eloundou et al., 2024). AI may automate some tasks, augment others, increase productivity, reorganize occupations, and alter pathways to professional advancement. Aggregate employment outcomes can mask unequal transitions. New opportunities may emerge even as particular routes to economic stability disappear; productivity gains need not flow equally to workers and owners.
The public-theological question is therefore: Who possesses sufficient transition capacity to move from technological disruption toward technological opportunity? Wealth, educational access, networks, geography, and retraining opportunities can affect the answer. Womanist analysis further asks whose poorly compensated, feminized, racialized, or otherwise invisible labor disappears inside narratives of efficiency. Churches should help communities prepare for transition without predicting outcomes with false certainty. Economic displacement is not ontological depreciation. A worker whose task loses market value has not become less human.
Data, Classification, and Cultural Extraction
Data become inputs for training, evaluation, prediction, personalization, and monetization, but not all information possesses the same moral status. Some is public, some personal, some relational, some culturally significant, and some sacred. Datafication requires institutions to select variables, define categories, and make aspects of life computationally legible. Black history supplies reason to scrutinize consequential classification, not proof that every contemporary algorithm replicates slavery or segregation. The relevant questions concern who defines categories, whether classifications can be challenged, and whether probabilistic judgments become institutional certainties. People may be represented by data without becoming reducible to data.
Black language, preaching, music, literature, oral tradition, and religious history may also become computational material. Generative systems can reproduce recognizable patterns in these traditions, sometimes supporting creativity and preservation, and sometimes creating extractive asymmetry. Cultural exchange is not exploitation. The justice questions concern who supplies the material, who governs reuse, who captures the resulting value, and what recourse exists when synthetic reproduction misrepresents people or traditions. Technological capability does not confer moral entitlement to cultural inheritance.
Surveillance and Asymmetric Visibility
The U.S. Senate’s 1976 investigation into federal surveillance and the disruption of civil-rights leadership offers reasons to scrutinize institutional monitoring, not automatic proof of wrongdoing by a present-day AI system. Scholarship on the surveillance of Blackness and racialized technological classification provides additional critical frameworks (Browne, 2015; Benjamin, 2019). Buolamwini and Gebru (2018) demonstrated disparities in the commercial gender-classification systems they tested; their findings must not be generalized to all facial-recognition systems. DMoPA names the concern asymmetric visibility: institutions may classify individuals computationally while those individuals cannot adequately understand, challenge, or correct consequential inferences. Prophetic Verification must examine claims made about people through institutional systems, not only claims made by people.
The Cloud Has a Geography
AI computation depends on material infrastructure (International Energy Agency, 2025; Shehabi et al., 2024; Smith et al., 2026). Data centers require electricity, cooling, transmission, land, and industrial supply chains. Their effects vary by energy mix, location, cooling technology, and design. The International Energy Agency (2025) documented significant global data-center electricity use and projected growth; those figures describe data centers generally, not AI workloads alone. No universal water-use-per-prompt figure adequately represents every facility. Independent technical evidence should establish material demands, while environmental-justice scholarship supplies the normative questions.
Bullard (1990) connects race and environmental burdens; Schlosberg (2007) emphasizes distribution, recognition, participation, and capabilities. NAACP (2026) advocacy concerning data-center development demonstrates that infrastructure is already a site of contestation but does not prove every facility harms Black communities. The proper questions concern benefits, burdens, public incentives, local participation, independent monitoring, and mechanisms of accountability. The moral object is not merely the chatbot; it is the sociotechnical arrangement.
Efficiency Does Not Determine Distribution
If AI enables an organization to produce more with fewer labor hours, technological capability does not determine whether the resulting value translates into higher wages, lower prices, greater profits, shareholder returns, or public revenue. Distribution is institutional. Technical success does not settle the justice question. Productivity gains can accrue to wages, lower prices, corporate profits, shareholders, or public revenue, depending on bargaining power, ownership, law, and organizational decisions. Communities may supply labor, data, or infrastructure without receiving proportional benefits. For Black public theology, the question is not only whether technology creates value, but who possesses the authority to allocate that value and to contest the terms of its production. This analysis also does not require technological fatalism: AI may expand research, accessibility, entrepreneurship, and the capacities of resource-constrained institutions. But opportunity is not self-executing. Communities need education, networks, capital, legal knowledge, governance literacy, and the capacity to negotiate. The question is how Black communities can participate in technological opportunity without accepting exploitation, dependency, or unequal burdens as its necessary price.
V. PROTECTION + PARTICIPATION + OWNERSHIP
From Technological Exposure to Black Institutional Agency
If AI operates within structures of labor, data, infrastructure, capital, and institutional power, critique alone is insufficient. Black public theology must move from identifying contradictions toward developing capacities to respond. Technological surrender adopts innovation because it seems inevitable, leaving questions of dignity and power for later. Technological retreat becomes so dominated by risk that communities withdraw from opportunities better-resourced institutions are using. Neither posture produces agency.
The reconstructive alternative is Protection + Participation + Ownership. These constituent ideas are not inventions of DMoPA; their contribution here is their integration within Black ecclesial technological ethics. Without protection, participation can become exploitation. Without participation, protection can become exclusion. Without ownership, participation can remain dependency.
Protection: The Capacity Not to Be Exploited
Protection requires more than access. Communities need epistemic safeguards against fabricated sources, synthetic media, and persuasive misinformation; data protections for personal, congregational, financial, pastoral, and archival information; economic preparation against fraud and poorly governed transitions; and civic protections where automated systems influence consequential decisions. Yet protection must not become fear. A community taught only what technology may do to it may remain poorly prepared to understand what technology enables it to do for itself.
Prophetic Verification therefore means protection without paranoia. Corporate claims of benefit, governmental claims of necessity, activist claims of harm, and Church claims of institutional responsibility should all remain accountable to evidence. The objective is to distinguish confidence from evidence.
Participation: Protection Cannot Become Exclusion
AI may lower barriers to capabilities concentrated in wealthy institutions. Students may receive educational assistance; small organizations may gain research, design, communication, and administrative capacity; accessibility tools may widen participation. Black technological ethics cannot train communities only to avoid while others prepare to create. Participation requires technological fluency and pathways into education, research, engineering, entrepreneurship, creative production, law, policy, governance, and institutional leadership.
Access asks whether people can use a system. Fluency asks whether they can use it effectively. Agency asks whether they can shape outcomes. Governance asks whether they influence rules. Ownership asks whether they participate in productive assets and value. These dimensions overlap but are not interchangeable. Prompt competence is useful; it is not technological self-determination.
Ownership: From Participation to Economic Capacity
Technological economies generate wealth through employment and ownership of businesses, intellectual property, applications, services, licensing arrangements, research enterprises, and productive assets. The question is whether Black communities will remain concentrated in consumption and labor or develop productive assets.
Ownership can increase bargaining power and wealth, but Black-owned systems can still exploit workers, misuse data, or violate workers' dignity. The triad is also not a complete political economy: labor protections, public regulation, taxation, competition policy, cooperative models, and democratic governance remain important.
Stewardship, Self-Determination, and Knowledge Production
Technological stewardship asks not merely whether a capability exists but whether, how, and toward what ends it should be used. Agency includes disciplined non-use: not every interaction should be automated, every dataset collected, every archive used for training, or every prediction turned into a decision. Eccher and Rossi (2025) develop technological self-determination as a framework for meaningful negotiating power within technological interdependence. This does not require technological separatism.
Knowledge production is also an act of agency. Black institutions should not remain merely populations studied by outside researchers. HBCUs, seminaries, churches, civil-rights organizations, professional associations, and research organizations can investigate AI use among clergy, AI and Black youth, technological trust, data governance, Womanist care ethics, and infrastructure burdens. The Justice Scholars Society can contribute interdisciplinary scholarship, verification, and public theology.
The Agency Test
When evaluating a technological arrangement, ask: What vulnerabilities does it create, and can we govern them? Who gains knowledge, opportunity, or decision-making power? Who retains assets, rights, or institutional capacity? Finally, what form of human flourishing does the arrangement order toward? A technology can be profitable yet dehumanizing, accessible yet manipulative, Black-owned yet exploitative, efficient yet unjust. Evaluate technological agency not merely by possession of power, but by what communities become through its exercise.
VI. THE BLACK CHURCH AS AN AI FORMATION INSTITUTION
An Ecclesial Ecosystem for Fluency, Verification, Agency, and Human Flourishing
If AI increasingly shapes the environments where Black communities learn, work, communicate, and encounter institutions, ecclesial engagement cannot end with Church technology policies. What capacities must communities develop to inhabit an AI-mediated society without surrendering theological discernment, historical consciousness, dignity, or agency?
This essay proposes that Black churches serve as AI formation institutions. The phrase names a function, not a new office or organizational model. It does not require every congregation to employ specialists or develop infrastructure. Churches vary in resources and mission, and technological sophistication does not measure ecclesial maturity. A small Church may have limited technical resources and extraordinary relational trust; a large Church may have sophisticated systems but inadequate accountability. The question is how faithfully the Church helps people interpret and exercise agency amid technological change.
From Digital Religion to AI Formation
Black Christian engagement with digital media did not begin with generative AI. Gault (2022) demonstrates Black religious identities and communities formed within and beyond physically located churches. Generative systems change the character of mediation by not only distributing content but also producing, synthesizing, personalizing, and interpreting it. That development requires formation, not mere access.
Seven capacities—Formation, Fluency, Verification, Protection, Participation, Ownership, and Governance—can be distributed across an ecclesial ecosystem.
Formation; Fluency and Verification
Formation begins with prior questions: What is intelligence? What is wisdom? What makes a person worthy? What should never be delegated? When does efficiency become morally insufficient? The Church begins not with what the machine can do but with the kind of human and communal life that technology should serve.
Communities also need sufficient technical fluency to distinguish among generative systems, recommendations, predictive analytics, and automated decisions; to understand why outputs may be false; and to recognize risks when information is shared on external platforms. Verification turns fluency into a practice of truthfulness: locate sources, test claims, assess provenance, identify contrary evidence, and acknowledge uncertainty. Prophetic speech cannot maintain credibility while abandoning evidentiary discipline.
Protection and Participation; Ownership and Governance
Protection prepares communities for synthetic fraud, privacy loss, misinformation, technological dependency, and occupational disruption. Participation ensures that protection does not become isolation. Black children should encounter AI not only as consumers but also as potential researchers, designers, critics, entrepreneurs, and governors of technology. Churches can connect people to HBCUs, community colleges, libraries, professional associations, and training networks rather than providing every skill themselves.
Ownership concerns pathways toward productive assets, intellectual property, research enterprises, and entrepreneurship where appropriate. Governance concerns rights, contracts, vendor relationships, data practices, public infrastructure, accountability, and appeal. Governance converts technical literacy into civic and institutional agency.
Intergenerational Formation and the Ecclesial Ecosystem
Technological change can widen generational distance, but churches can foster reciprocal learning. Younger members may contribute emerging technical familiarity, while elders may contribute institutional memory and experience with earlier social transformations. Neither form of knowledge is sufficient on its own.
The relevant unit is an ecclesial ecosystem: congregations contribute trust; denominations share resources; seminaries provide theological formation; HBCUs and universities conduct research; civil-rights organizations provide advocacy; attorneys provide legal expertise; entrepreneurs provide economic imagination; technologists provide technical understanding; and research organizations provide evidence and verification. DMoPA's De-Fragmentation names the reconstruction of relationships among forms of knowledge and responsibility whose separation weakens communal capacity. The goal is an accountable relationship, not erasing specialization.
Disciplined Non-Use and Human Flourishing
Mature technological formation includes the ability to refuse. Not every pastoral interaction should be automated, prayer requests made into data, archives used for training, or efficiency treated as mission. Sometimes the most consequential governance decision occurs before dependency develops.
The normative horizon is human flourishing, an established theological and interdisciplinary concept rather than a DMoPA invention (Charry, 2010; VanderWeele, 2017; Volf, 2016). Here it includes dignity, agency, relationship, economic possibility, historical consciousness, responsibility, participation, and freedom from domination. Innovation is not identical to flourishing. Efficiency is not identical to justice. Capability is not identical to wisdom.
Limitations and Research Agenda
This essay is constructive, not empirical. It does not establish causal effects of AI on Black religious formation. Algorithmic Formation and Performance–Formation Fragmentation are theoretical propositions requiring investigation. Black ecclesial life is heterogeneous across denomination, gender, sexuality, disability, age, class, geography, occupation, and institutional resources. The proposed capacities are not mandates for every congregation.
Future research should ask: How are Black clergy and congregants using generative AI, and how do they assess AI-generated theological authority? How does technological trust vary across generations, gender, class, disability, and ecclesial traditions? Under what conditions does AI-assisted sermon preparation strengthen or weaken exegetical judgment and ministerial formation? How do congregations govern sensitive data and negotiate vendor contracts? How do Black women experience AI-mediated labor, care, and hidden technological burdens? Who receives economic benefits from AI infrastructure, and who bears local energy, water, and environmental costs? Which partnerships among churches, HBCUs, technologists, and civil-rights organizations measurably strengthen community agency? These are research questions, not conclusions this essay establishes.
The Black Church need not become an AI industry. It can help build the relationships through which communities know enough, question enough, verify enough, participate enough, and organize enough to remain agents rather than merely subjects of technological change.
VII. CONCLUSION — THE ALGORITHM IS ALREADY TEACHING. IS THE CHURCH?
Formation, Dignity, Accountability, and Black Agency in the Age of Artificial Intelligence
Artificial intelligence poses Black ecclesial institutions with a challenge that extends beyond technological adoption. It raises questions about formation, dignity, accountability, and agency. AI should be understood not merely as a tool churches may use or a risk they must regulate, but as part of the environments through which people encounter knowledge, work, institutions, culture, and one another. This does not mean computational systems determine Black consciousness. Families, schools, churches, communities, economies, histories, cultures, and human agency remain formative. AI enters that ecology; it does not replace it.
Algorithmic Formation names computational participation in formative environments without reducing technological presence to technological determinism. AI can expand knowledge, accessibility, creativity, research, entrepreneurship, and institutional capacity while raising concerns about misinformation, surveillance, dependency, labor, data, and unequal distributions of power. The response is neither fear nor surrender. It is formation.
Dignity Without Competition
At the theological center stands the Non-Computability of Human Worth. Computational description, prediction, classification, productivity, or machine-comparable capability cannot exhaust a person's meaning. Human dignity cannot depend upon outperforming machines because human dignity never legitimately depended upon outperforming other human beings. A person may lose economic advantage without losing dignity. A prediction may describe probability without becoming destiny. A classification may inform a decision without exhausting the person classified. A human being remains more than the measure assigned to them.
Black theological traditions bring historical force to this proposition because Black communities have confronted systems that converted alleged differences in intelligence, race, gender, bodily capacity, legal status, and economic usefulness into hierarchies of human value. Womanist and disability critiques further expose the inadequacy of making dominant forms of capacity conditions of personhood. AI does not create that problem; it makes the problem newly visible.
Capability Without Abandoned Responsibility
The same boundary governs AI in ministry. It may assist with preaching, education, accessibility, pastoral resources, administration, and archives. The question is whether augmentation remains accountable. AI may participate in sermon production without automatically participating in preacher formation; generate relational language without inheriting the pastoral covenant; or synthesize theology without becoming neutral. Performance–Formation Fragmentation names a possible divergence between polished output and the human competence required to understand, evaluate, reproduce, and assume responsibility for it. That outcome is not inevitable. Responsible AI use can strengthen learning and judgment. Increasing technological capability must be accompanied by greater clarity about human and institutional responsibility.
Agency Beyond the Interface
Accountability requires tracing AI from the prompt to the power plant. Its material and institutional conditions include labor, data, capital, intellectual property, energy, water, infrastructure, and political authority. Protection + Participation + Ownership offers a reconstructive response: protection without participation can become exclusion; participation without protection can become exploitation; participation without meaningful ownership can remain dependency. Yet ownership itself is not justice. The normative horizon is human flourishing.
The Black Church's contribution may lie less in becoming technologically sophisticated than in expanding its capacity for formation and convening. No congregation can know everything about AI. Churches, HBCUs, seminaries, technologists, researchers, attorneys, entrepreneurs, civil-rights organizations, youth, and elders can bring different forms of knowledge into an accountable relationship.
This responsibility requires institutional humility. A Church cannot criticize opaque technological authority while protecting opaque ecclesial authority. Who decides? Who benefits? Whose voice is missing? Who can contest? Who remains accountable? These questions must move in both directions.
The Algorithm Is Already Teaching
The Black Church does not need to defeat the machine. It must help form human beings capable of deciding what machines are for. That requires distinguishing information from formation, intelligence from wisdom, prediction from destiny, simulation from covenant, productivity from dignity, access from agency, capability from authority, and innovation from justice.
Who will disciple Black people in the age of artificial intelligence? No algorithm should inherit formative authority by default. No corporation should possess it merely because its technology becomes ubiquitous. And no Church should assume it retains formative authority merely because it possesses a sanctuary.
Formation must be practiced. Dignity must be defended. Power must remain accountable. Agency must be built.
The crisis would be allowing increasing machine capability to coincide with decreasing human responsibility. The Black Church once helped Black people interpret a nation that repeatedly misinterpreted their humanity. The age of artificial intelligence presents another interpretive challenge—not a summons to preserve a world before AI, but an opportunity to help construct one in which technological power remains accountable to human dignity, communal agency, moral responsibility, and justice.
The algorithm is already teaching.
Is the Church?
REFERENCES
Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim code. Polity.
Browne, S. (2015). Dark matters: On the surveillance of Blackness. Duke University Press.
Bullard, R. D. (1990). Dumping in Dixie: Race, class, and environmental quality. Westview Press.
Buolamwini, J., & Gebru, T. (2018). Gender shades: Intersectional accuracy disparities in commercial gender classification. Proceedings of Machine Learning Research, 81, 77–91. https://proceedings.mlr.press/v81/buolamwini18a.html
Butler, P. (2024). Black theology × artificial intelligence. In B. Singler & F. Watts (Eds.), The Cambridge companion to religion and artificial intelligence (pp. 182–200). Cambridge University Press. https://doi.org/10.1017/9781009031721.013
Cannon, K. G. (1988). Black womanist ethics. Scholars Press.
Caswell, M., Migoni, A. A., Geraci, N., & Cifor, M. (2017). ‘To be able to imagine otherwise’: Community archives and the importance of representation. Archives and Records, 38(1), 5–26. https://doi.org/10.1080/23257962.2016.1260445
Charry, E. T. (2010). God and the art of happiness. Eerdmans.
Cone, J. H. (1970). A black theology of liberation. Lippincott.
Cone, J. H. (1975). God of the oppressed. Seabury Press.
Douglass, F. (1852). Oration, delivered in Corinthian Hall, Rochester, by Frederick Douglass, July 5th, 1852. Lee, Mann & Co. https://archive.org/details/orationdelivered00fred
Douglass, F. (1866). Reconstruction. The Atlantic Monthly, 18(110), 761–765.
Eccher, L., & Rossi, U. (2025). Beyond technological sovereignty: Practising technological self-determination in platform cooperativism. Environment and Planning A: Economy and Space, 57(4), 385–401. https://doi.org/10.1177/0308518X251325743
Eiesland, N. L. (1994). The disabled God: Toward a liberatory theology of disability. Abingdon Press.
Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2024). GPTs are GPTs: Labor market impact potential of LLMs. Science, 384(6702), 1306–1308. https://doi.org/10.1126/science.adj0998
Gault, E. D. (2022). Networking the Black Church: Digital Black Christians and hip hop. New York University Press.
Gottfried, J., Bishop, W., Anderson, M., Faverio, M., Park, E., & McClain, C. (2026, June 17). Racial and ethnic differences in how adults use and view AI. Pew Research Center. https://www.pewresearch.org/internet/2026/06/17/racial-and-ethnic-differences-in-how-adults-use-and-view-ai/
Henderson, M. D. (2026). Sanctified spaces: Integrating ambient intelligence and smart environments into U.S. Black Baptist churches. Humanities and Social Sciences Communications, 13, Article 397. https://doi.org/10.1057/s41599-026-06756-5
International Energy Agency. (2025). Energy and AI. https://www.iea.org/reports/energy-and-ai
LaRue, C. J. (2000). The heart of Black preaching. Westminster John Knox Press.
Lincoln, C. E., & Mamiya, L. H. (1990). The Black Church in the African American experience. Duke University Press.
Mitchell, H. H. (1990). Black preaching: The recovery of a powerful art. Abingdon Press.
NAACP. (2026). NAACP’s 2026 recommendations for protecting frontline communities to stop dirty data. https://naacp.org/sites/default/files/documents/Dirty%20Data%202026%20Recs.pdf
Parasuraman, R., & Manzey, D. H. (2010). Complacency and bias in human use of automation: An attentional integration. Human Factors, 52(3), 381–410. https://doi.org/10.1177/0018720810376055
Reynolds, T. E. (2008). Vulnerable communion: A theology of disability and hospitality. Brazos Press.
Risko, E. F., & Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9), 676–688. https://doi.org/10.1016/j.tics.2016.07.002
Sande, N. (2026). Algorithmic empathy: Assessing AI’s capacity for presence in pastoral care. Pastoral Psychology. https://doi.org/10.1007/s11089-026-01340-9
Schlosberg, D. (2007). Defining environmental justice: Theories, movements, and nature. Oxford University Press.
Shehabi, A., Smith, S. J., Hubbard, A., Newkirk, A., Lei, N., Siddik, M. A. B., Holecek, B., Koomey, J. G., Masanet, E. R., & Sartor, D. A. (2024). 2024 United States data center energy usage report. Lawrence Berkeley National Laboratory. https://doi.org/10.71468/P1WC7Q
Smith, S. J., Hubbard, A., Newkirk, A., Ganeshalingam, M., Holecek, B., Sartor, D. A., Mills, M., & Shehabi, A. (2026). United States data center energy usage report: 2025 update. Lawrence Berkeley National Laboratory. https://doi.org/10.71468/P1RP4F
Thomas, F. A. (2013). They like to never quit praisin’ God: The role of celebration in preaching (Rev. & updated ed.). Pilgrim Press.
Tretter, M., & Glatzel, M. (2026). Not human enough? Rethinking AI’s place in pastoral care. Pastoral Psychology. https://doi.org/10.1007/s11089-026-01351-6
U.S. Copyright Office. (2024). Copyright and artificial intelligence, part 1: Digital replicas. https://www.copyright.gov/ai/Copyright-and-Artificial-Intelligence-Part-1-Digital-Replicas-Report.pdf
U.S. Copyright Office. (2025). Copyright and artificial intelligence, part 3: Generative AI training (Pre-publication version). https://www.copyright.gov/ai/Copyright-and-Artificial-Intelligence-Part-3-Generative-AI-Training-Report-Pre-Publication-Version.pdf
U.S. Senate, Select Committee to Study Governmental Operations with Respect to Intelligence Activities. (1976). Intelligence activities and the rights of Americans (Book II). U.S. Government Printing Office.
VanderWeele, T. J. (2017). On the promotion of human flourishing. Proceedings of the National Academy of Sciences, 114(31), 8148–8156. https://doi.org/10.1073/pnas.1702996114
Volf, M. (2016). Flourishing: Why we need religion in a globalized world. Yale University Press.
Williams, D. S. (1993). Sisters in the wilderness: The challenge of womanist God-talk. Orbis Books.
Yong, A. (2007). Theology and Down syndrome: Reimagining disability in late modernity. Baylor University Press.
