Photo Courtesy: Image by Gerd Altmann from Pixabay | For representational purpose only
Dr Wungtei Buchem
Associate Professor, Trinity Theological College, Thahekhu, Dimapur
The rapid advancement of Artificial Intelligence (AI) has fundamentally transformed contemporary society. Generative AI systems now write essays, translate languages, diagnose diseases, generate computer code, compose music, and increasingly influence public policy and economic decision-making. Governments describe AI as a defining technology of the twenty-first century, while multinational corporations invest billions to secure technological leadership. Yet its benefits are unevenly distributed. Most frontier AI models are developed by corporations headquartered in the United States, while China has emerged as its principal geopolitical competitor. Much of Africa, Latin America, and Asia functions primarily as a consumer rather than producer of advanced AI technologies. Consequently, AI development reflects significant asymmetries in technological capacity, data ownership, computational resources, and regulatory influence.
This reality raises a fundamental question: Does AI represent merely technological progress, or does it also constitute a new form of colonial power? This article argues that AI should not be understood simply as software or computational intelligence. Rather, it increasingly operates within global political economies that shape whose knowledge is collected, whose languages are prioritized, whose values are encoded, and whose voices remain marginalized. In this respect, AI may reproduce forms of domination analogous to earlier colonial structures, not through territorial occupation but through control over data, digital infrastructure, computational resources, and epistemic authority.
The concept of digital colonialism has therefore become increasingly important within contemporary scholarship. Unlike classical colonialism, which relied upon military conquest and territorial occupation, digital colonialism operates through digital platforms, algorithmic governance, cloud infrastructures, surveillance technologies, and data extraction. Colonial power no longer requires direct political administration; it increasingly operates through technological dependency and informational asymmetry. For scholars in the Global South, this raises pressing ethical and theological questions. How might indigenous knowledge traditions survive within AI systems largely trained on dominant languages and datasets? Can AI perpetuate epistemic injustice by privileging Western knowledge while marginalizing oral cultures and local epistemologies? How should theology respond when technological power becomes concentrated among a few global actors?
These questions are particularly significant for postcolonial biblical interpretation, which seeks to expose structures of domination embedded within both ancient texts and contemporary societies. If colonialism historically involved the control of land, labor, and knowledge, digital colonialism extends these dynamics into cyberspace through the governance of information, algorithms, and AI.
The concept of digital colonialism has emerged from broader discussions of globalization, surveillance capitalism, and the political economy of data. Nick Couldry and Ulises A. Mejias discuss the emergence of data colonialism, whereby human life is appropriated through continuous extraction of behavioral data for economic value. They contend that data relations increasingly resemble colonial relations because both depend upon appropriating resources from populations without equitable participation in decision-making or ownership (Couldry and Mejias, 2019). Kate Crawford demonstrates that AI systems depend upon extensive material infrastructures, including mining, global labor, energy consumption, and data extraction. Contrary to representations of AI as immaterial intelligence, Crawford argues that AI is deeply embedded within historical networks of economic inequality and environmental exploitation (Crawford, 2021). Amar Ahmad, Yvonne Vallès, and Youssef Idaghdour further illustrate how algorithmic systems can reproduce racial hierarchies rather than eliminate them. Technologies often inherit social prejudices embedded within training data, reinforcing structural inequalities under the appearance of computational neutrality (Ahmad, Vallès, Idaghdour, 2026).
From a postcolonial perspective, Edward Said's seminal work Orientalism remains foundational. Said demonstrates that knowledge production itself constitutes a mechanism of imperial power. Colonial domination required not only military conquest but also systems of representation through which colonized peoples were defined and governed (Said, 1978). AI similarly raises questions concerning who constructs digital representations of humanity and whose epistemologies become normative. Gayatri C. Spivak's question, "Can the Subaltern Speak?" remains profoundly relevant in the age of AI. If algorithms are trained primarily on dominant linguistic and cultural datasets, marginalized communities risk becoming digitally invisible or misrepresented (Spivak, 1988). Homi K. Bhabha complicates simplistic understandings of colonial domination by emphasizing hybridity, mimicry, and cultural negotiation. His insights encourage scholars to view AI not merely as Western domination but as a contested space where local communities negotiate, adapt, resist, and transform technological systems (Bhabha, 1994).
Within biblical studies, R. S. Sugirtharajah, Fernando F. Segovia, Musa W. Dube, and Uriah Y. Kim have developed postcolonial hermeneutics as a means of exposing imperial ideologies embedded within biblical interpretation while retrieving marginalized voices (Sugirtharaj, 2001). Their work provides valuable methodological resources for theological engagement with contemporary digital empires.
Postcolonial theory emerged as a critical response to the enduring consequences of European colonialism. Although formal empires largely disappeared during the twentieth century, colonial patterns of economic dependency, cultural domination, and epistemic inequality continue to shape global relations. Edward Said argued that empire operates through systems of knowledge as much as through military force. Colonial discourse constructs categories of civilization and backwardness that legitimize unequal power relationships. Building upon Said, Walter D. Mignolo and Aníbal Quijano describe this continuing phenomenon as the coloniality of power, whereby colonial structures survive beyond colonial administration through enduring hierarchies of race, knowledge, economy, and governance (Mignolo and Quijano, 2011).
This framework provides an important lens for analyzing AI. AI is frequently portrayed as objective and universal, yet every system reflects choices concerning data collection, language prioritization, computational infrastructure, funding, regulation, and ethical values. These choices are never entirely politically neutral. Consequently, AI should be understood not merely as technological innovation but as a site where power, knowledge, and ideology intersect. A postcolonial analysis therefore asks not simply what AI can do, but whose interests it serves, whose knowledge it privileges, and whose voices remain excluded.
Classical colonialism depended upon territorial conquest, military occupation, economic exploitation, and cultural assimilation. European empires justified expansion through narratives of civilization, modernization, and religious mission while extracting labor, natural resources, and indigenous knowledge. Although political colonialism formally ended in much of the twentieth century, its structures can persist through new economic and technological formations. Mignolo therefore distinguishes colonialism as a historical political system from coloniality as an enduring logic of domination that survives beyond formal empire.
Unlike nineteenth-century empires, contemporary technological powers rarely annex territory. Instead, they exercise influence through digital infrastructures, cloud computing, data ownership, software ecosystems, and algorithmic governance. Power increasingly resides not in controlling colonies but in controlling information, computational capacity, and digital standards. AI must be understood within this transformation. Couldry and Mejias describe this process as data colonialism. Contemporary digital corporations appropriate human experience as a resource, much as colonial empires appropriated land and labor. Search queries, online purchases, biometric scans, GPS locations, and social media interactions become raw material for algorithmic production. Human life is increasingly transformed into data that can be extracted, commodified, and monetized.
This comparison should not suggest that digital technologies replicate colonialism in every respect. Historical colonialism involved conquest, slavery, and profound physical violence that cannot simply be equated with contemporary technological practices. Nevertheless, the analogy illuminates structural similarities involving extraction, dependency, asymmetrical power, and unequal participation in decision-making. Colonialism converted land into economic capital; digital colonialism increasingly converts human behavior into informational capital. Consequently, AI functions not merely as software but as an infrastructure of political economy. Whoever controls data, computational resources, and algorithmic innovation exercises considerable influence over global knowledge production and economic development.
AI depends fundamentally upon data. Machine-learning systems acquire predictive capabilities by processing enormous quantities of text, images, audio, and behavioral information. Data has therefore become one of the digital age's most valuable strategic resources. Unlike traditional natural resources, however, data is produced through everyday human activity. Individuals generate data through smartphones, digital payments, online education, transportation systems, health records, and social media. Users often receive digital services without directly paying monetary fees; instead, they exchange behavioral information that fuels algorithmic learning and targeted advertising.
Shoshana Zuboff argues that surveillance capitalism transformed personal experience into “behavioral surplus,” allowing corporations to predict and influence future human behavior (Zuboff, 2020). Such predictive capacities generate enormous commercial value while concentrating informational power within a limited number of corporations. Mahmut Kandemir further demonstrates that AI systems remain deeply material despite their digital appearance. AI requires rare earth minerals, massive energy consumption, data centers, underpaid data-labeling labor, global supply chains, and extensive computational infrastructure (Kandemir, 2025). AI therefore cannot be understood apart from global political economy.
From a postcolonial perspective, an important question emerges: Who owns the world's data? Much of the Global South generates extensive digital data while possessing comparatively limited capacity to develop large-scale AI systems. Developing nations can therefore become suppliers of data while remaining dependent upon technologies developed elsewhere. This imbalance mirrors earlier colonial economic structures in which colonies exported raw materials while importing finished goods. The digital economy risks reproducing a similar pattern: the Global South supplies data, technological powers develop AI models, economic value accumulates within technologically advanced states, and developing nations become consumers of AI services. This asymmetry raises profound questions concerning digital sovereignty and technological justice.
The twenty-first century has witnessed unprecedented geopolitical competition concerning AI. The United States presently maintains leadership in frontier AI research through corporations such as OpenAI, Google DeepMind, Anthropic, Microsoft, Meta, and others. These organizations possess extraordinary computational resources, specialized research personnel, and access to enormous datasets.
Meanwhile, China has invested heavily in AI through state-directed industrial policy. Chinese firms including Baidu, Alibaba, Tencent, Huawei, and more recently DeepSeek have expanded rapidly under government initiatives aimed at technological self-sufficiency and global competitiveness. China's Digital Silk Road similarly extends digital infrastructure into numerous countries through telecommunications, cloud services, smart-city technologies, and AI applications. This rivalry increasingly resembles a struggle for technological sovereignty. AI has become central not only to economic development but also to military strategy, cybersecurity, public administration, and international influence.
However, the postcolonial question differs from conventional geopolitical analysis. Rather than asking which superpower will dominate AI, postcolonial scholarship asks what becomes of societies that possess neither American technological infrastructure nor Chinese computational capacity. The Global South risks becoming an arena where competing technological powers expand influence through digital platforms, infrastructure investment, cloud services, and AI governance models. Countries may acquire advanced technologies without developing corresponding technological autonomy.
This situation resembles earlier colonial patterns in which local economies became integrated into global systems over which they exercised limited control. Nevertheless, caution is necessary. The United States and China represent different political systems, regulatory approaches, and economic models. A postcolonial analysis should therefore avoid reducing both countries to identical forms of digital empire. The concern lies instead in structural dependence rather than simplistic moral equivalence.
Perhaps the most significant postcolonial challenge concerns knowledge. Said demonstrated that colonialism required systems of representation that defined colonized peoples through categories constructed by imperial centers (Said, 1983). Knowledge itself became an instrument of domination. AI raises similar concerns because AI systems learn from existing datasets. If particular languages, cultures, histories, or epistemologies remain underrepresented, AI outputs may systematically privilege dominant perspectives.
English remains the dominant language of global AI development. Although multilingual models have expanded considerably, many indigenous languages remain digitally marginalized. Oral traditions, local histories, tribal knowledge systems, and community-based epistemologies often possess limited digital representation. Consequently, AI may reproduce what Boaventura de Sousa Santos calls epistemicide: the destruction or marginalization of alternative knowledge systems through dominant forms of modern knowledge. Her book Epistemologies of the South: Justice Against Epistemicide explores “cognitive injustice,” the failure to recognize different ways of knowing by which people across the globe understand and give meaning to their existence. Santos shows why global social justice is impossible without global cognitive justice and argues for recovering and valuing the world's epistemological diversity (Santos, 2014).
Spivak's question, "Can the Subaltern Speak?" acquires renewed significance in this digital context. The issue is no longer merely whether marginalized communities possess political voice but whether they possess sufficient digital presence to shape algorithmic knowledge production. When AI systems answer questions about history, religion, medicine, culture, or ethics primarily through dominant datasets, marginalized communities may become digitally invisible. Their knowledge exists but remains computationally inaccessible. For indigenous peoples throughout Asia, Africa, Latin America, and Oceania, this raises serious concerns regarding cultural preservation. Languages with limited digital corpora risk exclusion from future AI systems, reinforcing inequalities in education, communication, and technological participation.
AI is frequently described as objective because algorithms rely upon mathematical computation. However, numerous scholars have demonstrated that algorithmic systems can reproduce existing social inequalities. Ruha Benjamin discusses the relationship between machine bias and systemic racism, analyzing cases of “discriminatory design” and proposing a socially conscious approach to technological development. In Race After Technology (2019), Benjamin demonstrates how emerging technologies can reinforce White supremacy and deepen social inequity by encoding and amplifying racial hierarchies, ignoring social divisions, or attempting to correct racial bias in ways that reproduce it.
Mike Zajko similarly argues that technological systems can encode historical forms of racial discrimination while presenting themselves as neutral and scientific. For AI developers, the evident unfairness of algorithmic technologies, including racist or sexist outputs, has created conceptual difficulties for which they were often unprepared. When society is deeply and structurally unequal and technologies are part of those structures, the question becomes whether algorithms should reproduce inequality or work to change it (Zajko, 2022).
Safiya Umoja Noble likewise demonstrates that search engines have historically reflected racialized and gendered biases rather than objective knowledge (Noble, 2018). Her book Algorithms of Oppression examines the power of algorithms in the age of neoliberalism and how digital decisions reinforce oppressive social relationships and new modes of racial profiling, which she terms technological redlining. By revealing how capital, race, and gender contribute to unequal conditions, Noble exposes technological redlining as an expanding form of digital inequality.
From a postcolonial perspective, algorithmic bias should therefore be understood not merely as a technical error but as an expression of deeper structures of coloniality. Algorithms inherit assumptions embedded within training data; training data reflect historical inequalities; and historical inequalities can influence technological outputs. Consequently, AI may reproduce existing hierarchies unless deliberate corrective measures are implemented. This phenomenon resembles colonial archives, which often claimed administrative neutrality while privileging imperial perspectives and marginalizing indigenous voices. AI training datasets risk functioning similarly when dominant cultures overwhelmingly determine the informational universe from which algorithms learn. AI should therefore not be imagined as existing outside history. It is another site where struggles concerning representation, authority, and knowledge continue to unfold.
Postcolonial biblical hermeneutics has consistently argued that biblical texts emerged within complex contexts of empire, colonization, resistance, and identity formation. Rather than treating Scripture as politically neutral, postcolonial interpreters investigate how imperial ideologies shape biblical narratives and their later interpretations. Sugirtharajah argues that biblical interpretation should expose colonial assumptions while retrieving marginalized voices frequently silenced within dominant readings (Sugirtharajah, 2001). Similarly, Fernando F. Segovia insists that every interpretation emerges from a social location and therefore reflects particular political and cultural interests (Segovia, 2000).
Biblical interpretation is never detached from power. This methodological insight provides an important framework for theological engagement with AI. AI systems likewise appear objective and neutral, yet every algorithm reflects decisions regarding language, data selection, computational priorities, and institutional values. Just as postcolonial hermeneutics interrogates ideological assumptions embedded within biblical interpretation, it must also interrogate assumptions embedded within AI technologies. AI therefore becomes another “text” requiring critical interpretation. Theological reflection should not ask merely whether AI functions efficiently but also: Who benefits? Who is represented? Who remains invisible? Who controls technological authority? Whose knowledge becomes normative? These questions parallel those raised by postcolonial biblical scholars concerning empire, Scripture, and interpretation.
The story of the Tower of Babel offers a provocative theological metaphor for the concentration of political and economic power that suppresses genuine human diversity. Today's digital infrastructure increasingly exhibits similar tendencies. A relatively small number of corporations control cloud computing, advanced semiconductor design, frontier AI models, search engines, operating systems, and global digital communication. The issue is not technological innovation itself but technological systems that become instruments of centralized domination. God's judgment upon Babel does not condemn human creativity; rather, it interrupts the monopolization of human power. Babel may therefore be interpreted as a theological critique of technological centralization rather than technological development itself.
Current research in AI and hermeneutics asks whether AI's interpretive authority and data-driven insights might overshadow human traditions and introduce biases. AI enhances text processing and accessibility but also raises concerns about interpretive integrity. Some research findings indicate that “AI enhanced interpretive capacity, broadened accessibility, and preserved historical context, but also posed risks, including potential biases, ethical concerns, and challenges to interpretive authority” (Nelson Njok, Chinelo Nelson and Chukwuma-Offor, 2025). While AI can enrich biblical interpretation, it should remain supplementary and respect theological and ethical principles. Responsible collaboration requires ethical guidelines, diverse perspectives, accessibility, spiritual depth, and strict data privacy in religious archives.
The implications of digital colonialism become particularly visible in the Asia-Pacific region. Many indigenous communities possess rich oral traditions, ecological knowledge, linguistic diversity, and communal systems of memory that remain underrepresented within global digital infrastructures. AI systems trained predominantly on English language datasets may fail to recognize these epistemologies. The consequence is not merely technical exclusion but epistemological marginalization.
Postcolonial biblical interpreters have long argued that colonialism marginalized indigenous knowledge by privileging Western epistemological frameworks. Digital technologies risk repeating this pattern when local communities participate only as consumers rather than contributors to AI development. This concern extends to tribal communities throughout Northeast India. Indigenous languages (Mashungam and Khamrang, 2025) with limited digital corpora remain vulnerable within future AI ecosystems. If these languages are absent from digital archives, future AI systems will have limited capacity to preserve or transmit them. Digital exclusion can therefore become cultural exclusion.
A postcolonial theology of AI calls not merely for technological access but for epistemic justice. Local languages should become AI languages; indigenous histories should become digital histories; and community knowledge should become computationally visible. Only then can AI contribute to genuine human flourishing rather than reinforce historical inequalities.
Postcolonial scholars have consistently emphasized that colonialism was not merely territorial but epistemological. Colonized peoples were frequently taught that legitimate knowledge originated elsewhere. AI risks reproducing similar assumptions if computational authority becomes concentrated within a small number of institutions. A decolonial AI must therefore recognize multiple forms of knowing. Scientific knowledge remains indispensable, but ecological wisdom, oral tradition, community memory, indigenous medicine, and local ethical traditions likewise constitute valuable forms of human knowledge. AI systems designed for genuinely global communities should reflect this epistemological plurality rather than assume the universality of a single cultural framework.
Technological governance likewise requires democratization. Decisions concerning AI safety, regulation, model deployment, and digital standards remain concentrated within governments and corporations possessing significant technological resources. Developing nations frequently participate only after major technological architectures have been established. A postcolonial framework therefore advocates greater Global South participation in international AI governance. Such participation should extend beyond political representation to include theologians, philosophers, indigenous leaders, linguists, ethicists, educators, and civil society organizations. AI affects humanity as a whole; its governance should therefore reflect humanity as a whole.
The preceding discussion suggests that theology should neither reject nor romanticize AI. Christian theology may contribute several ethical principles. First, AI should serve human dignity rather than market domination. Second, technological development should promote justice rather than dependency. Third, digital infrastructures should protect cultural plurality rather than encourage epistemological uniformity. Fourth, technological governance should include historically marginalized voices. Finally, wisdom must remain prior to intelligence.
The biblical tradition consistently evaluates political and technological power according to justice. The measure of AI therefore cannot simply be computational performance. Its ultimate ethical measure is whether it promotes the flourishing of all peoples, especially those historically excluded from structures of power.
If AI can reproduce structures of digital colonialism, postcolonial theology cannot remain merely diagnostic. It must also become constructive. The challenge is not simply to criticize technological concentration but to imagine ethical alternatives that promote justice, plurality, and human dignity. Mignolo reminds us that decoloniality is not merely the rejection of Western modernity but the recovery of epistemic plurality and recognition that no single civilization possesses a monopoly on truth or knowledge. Applied to AI, this calls for systems that are technically sophisticated but also culturally diverse, linguistically inclusive, and ethically accountable.
Current frontier AI models are trained predominantly on datasets generated from digitally dominant societies. Consequently, the histories, languages, and intellectual traditions of many indigenous peoples remain underrepresented. A decolonial AI therefore requires intentional investment in multilingual corpora, indigenous archives, and community-led data initiatives. This process should not become another form of extraction in which local communities merely provide data for external institutions. Indigenous peoples should retain meaningful participation in decisions concerning data ownership, governance, access, and benefit-sharing. Data sovereignty is therefore central. Indigenous communities increasingly argue that data concerning their histories, cultures, and identities should remain under collective stewardship. Such an approach challenges extractive models of AI development while promoting technological justice.
The future of AI should not become another chapter in global inequality. Many countries in Africa, Asia, Latin America, and the Pacific possess vibrant intellectual traditions yet remain technologically dependent upon infrastructures developed elsewhere. Postcolonial theology therefore calls for a transition from technological dependency toward technological participation. Universities in the Global South should invest in AI research rooted in local realities. Governments should encourage linguistic diversity within digital technologies. Churches should participate actively in AI ethics rather than assuming that technological questions belong exclusively to computer scientists. A call to digital adaptability becomes meaningful only when adaptability does not come at the cost of identity and participation. Otherwise, adaptability risks becoming a surrender to the very hegemony it seeks to overcome. Particularly within Asia-Pacific contexts, indigenous communities possess sophisticated ecological knowledge, communal ethics, and traditions of relational responsibility that may enrich global AI ethics. Rather than treating these traditions as historical curiosities, decolonial AI should recognize them as indispensable intellectual resources for humanity's technological future.
To conclude, AI should be understood not merely as technological innovation but as a significant site of political, economic, and epistemological power. While AI offers extraordinary opportunities for scientific advancement and human flourishing, it can also reproduce historical inequalities through digital infrastructures. Employing postcolonial theory, this study has argued that contemporary AI development exhibits important structural similarities to earlier forms of coloniality, particularly through data extraction, epistemic domination, technological dependency, and concentrated digital power.
The rise of digitally powerful nations demonstrates that AI increasingly functions within broader geopolitical competition. Yet from a postcolonial perspective, the central question is not which superpower prevails but whether the Global South participates as an equal partner in shaping humanity's technological future. Ultimately, AI’s future should not be determined solely by computational capability. Its deepest ethical question concerns whether technological progress advances the common good while respecting the dignity, diversity, and voices of all peoples. From a postcolonial theological perspective, genuine innovation is measured not only by what machines can accomplish but also by whether technological systems promote justice, preserve cultural plurality, and enable historically marginalized communities to participate fully in shaping the digital future. AI need not become another empire. It may instead become a shared human resource if governed according to justice, humility, solidarity, and epistemic plurality.
