Introduction
If colonialism once marched into Africa wearing khaki, clutching Bibles in one hand and British flags in the other, today’s invaders arrive cloaked in sleek devices and digital dashboards, speaking the gospel of Artificial Intelligence and algorithmic efficiency. Our forebears were herded into indirect rule by colonial administrators who claimed to bring “civilization” but delivered control. Today, we are being seduced into a new digital hegemony, where foreign algorithms determine everything from who gets a loan in Kenya’s mobile lending frenzy, to who gets stopped by facial recognition cameras piloted in parts of South Africa, to how disinformation spreads faster in Nigerian elections than INEC can count ballots. Back then, it was the Royal Niger Company extracting palm oil and rubber in exchange for trinkets; now it’s Silicon Valley giants extracting biometric data, behavioral patterns, and voice prints in exchange for free apps and predictive policing.
For instance, in Lagos, surveillance partnerships with Chinese tech firms are mushrooming under the guise of “smart city” solutions, never mind that no Nigerian voted for this digital dragnet. In Rwanda, drones deliver medicine while data on citizen movements are quietly archived for “security.” Across the continent, governments are quietly handing over governance to machine logics they neither understand nor control, outsourcing political responsibility to a faceless army of algorithms coded by cultural outsiders. The social contract is being rewritten not in parliaments, but in the Terms & Conditions we never read.
So, the question is both timely and terrifying: Are African states entering a new Faustian bargain by signing a social contract not with their citizens, but with the unseen Pharaohs of Silicon Valley? And if so, what becomes of the citizen is now reduced to a data point, a probability score, an algorithmic subject?
Historical Context: From British Indirect Rule to Silicon Indirect Control
Social contract theory, in its classic Hobbesian, Lockean, or Rousseauian form, is grounded in the principle of mutual consent between the governors and the governed, a voluntary surrender of some freedoms in exchange for protection, justice, and order. But colonial Africa was denied this philosophical luxury; its so-called “contract” was imposed, not negotiated. British indirect rule, particularly as perfected by Lord Lugard in Northern Nigeria, transformed indigenous institutions into administrative extensions of colonial authority. Back then, colonial power wore a native mask, where emirs were retained, not out of respect, but as instruments of remote control, an arrangement which historian Mahmood Mamdani calls the “bifurcated state”: authoritarian rule at the local level, and liberalism at the center for colonialists.
Fast-forward to today, and we find a disturbingly similar template. The Big Tech companies such as Facebook (now Meta), Google, Huawei and Palantir provide the infrastructure, while African governments, like colonial chiefs of old, serve as compliant facilitators. Consider Nigeria’s partnership with the World Bank to integrate biometric IDs under the National Identity Management Commission (NIMC), a project so shrouded in opacity that most Nigerians don’t even know their data is being used to train facial recognition tools abroad (Privacy International, 2020).
Where Lugard used the native authority to subdue the natives, the new Pharaohs use AI tools to manage populations—predicting crime before it happens, suppressing dissent before it materializes, and influencing elections before the first ballot is cast.
The New Algorithmic Order: Government by Proxy
AI in Africa is not inherently evil, its applications in healthcare, such as Zipline’s medical drone deliveries in Rwanda and Ghana, alongside precision agriculture tools in Kenya and mobile credit scoring systems, reveal its immense potential to address development gaps. However, potential without democratic oversight becomes a Trojan horse with an American accent: it promises efficiency while importing foreign assumptions and power asymmetries. At the heart of AI lies the mathematics of generalization, algorithms trained on historical data to predict future outcomes. Yet, when these models are deployed in African socio-economic contexts without proper localization, they risk reproducing the very biases, exclusions, and epistemic blind spots embedded in their original datasets and design assumptions. In such cases, rather than solving problems, AI quietly amplifies them but just more efficiently.
Let us take mobile lending algorithms in East Africa, apps like Tala and Branch use AI to assess creditworthiness based on text messages, call logs, and geolocation, not financial history. A 2021 study by FSD Kenya revealed that over 30% of borrowers were blacklisted for defaulting on microloans, sometimes due to inaccurate algorithmic assessments. People have lost their livelihoods because a bot misread their digital behavior.
Even more chilling is the partnership between Uganda and Huawei in surveillance deployment. In 2019, Uganda used Chinese AI-powered facial recognition to track and arrest political protesters, particularly supporters of Bobi Wine. The state, aided by foreign tech, crushed dissent before it could march to the ballot box (The Wall Street Journal, 2019). Here, AI doesn’t serve the citizen, it serves power.
Digital Neo-Colonialism: From Data as Oil to Data as Chains
In the colonial economy, raw materials were extracted and exported; in today’s algorithmic economy, it is raw identity that fuels global systems. African faces, voices, fingerprints, geolocation data, and browsing behavior, anything that can be digitized is harvested to train AI models in the Global North. Just as British companies once shipped palm oil from the Niger Delta to Liverpool, today’s African data silently powers surveillance architectures, financial models, and advertising engines from California to Shenzhen. But unlike oil, identity is non-renewable, once compromised, it is forever vulnerable. And yet, this extraction continues with minimal resistance, often framed as “development” or “digital inclusion.”
Beneath the sleek promise of innovation lies a fundamental flaw: AI systems do not possess moral reasoning; they are built to optimize for statistical efficiency, not justice. Governance, however, cannot be reduced to an equation, if a predictive policing algorithm disproportionately targets youth in Mushin, Lagos or Kibera, Nairobi, based on flawed or biased historical data, the harm is not accidental, it is systemic. It doesn’t merely misclassify individuals; it reinscribes structural inequality into code, making discrimination faster, quieter, and harder to challenge.
This is how data colonialism manifests in the 21st century, not through gunboats or treaties, but through invisible scripts and silent software updates. African digital subjects become mere inputs in decision-making systems governed by foreign logics and commercial interests. The erosion of sovereignty no longer requires conquest or coups; it now creeps in through unregulated APIs, opaque partnerships, and the outsourcing of state functions to machine learning systems whose accountability mechanisms are, at best, elusive. In this quiet capture of identity and agency, we see not the end of colonization—but its digital mutation.
Toward an Afro-Centric Social Contract for AI
The original idea of the social contract, whether in Hobbesian or Rousseauian terms was based on a clear principle: citizens surrender certain freedoms to the state in exchange for protection, justice, and accountable governance. But today, that responsibility is quietly being outsourced—not to elected representatives or trusted public institutions, but to algorithms and private technology firms. This shift raises urgent questions: how can people hold systems accountable when they cannot see, understand, or challenge the logic behind them? The answer is not to reject Artificial Intelligence, but to root it in African realities by training it on local languages, guiding it with our cultural values, and regulating it through democratic norms. AI must reflect our traditions of collective decision-making, communal responsibility, and consensus-building, not override them. Only then can it serve as a tool of empowerment, rather than a mechanism of silent control.
To reach that goal, Africa must go beyond passive adoption and actively shape the future of its digital governance. We must resist algorithmic rule without representation. We need what philosopher Ruha Benjamin calls “sociotechnical imaginaries”, which is visions of tech rooted in justice, culture, and history. Afro-centric AI must not be merely about “inclusion” in existing systems; it must mean rewriting the system itself to reflect African values: communalism, consent, justice, and dignity.
Imagine AI systems developed in Yoruba or Swahili, trained on locally sourced data, and governed by public boards with technical, legal, and civic expertise. Imagine algorithmic tools for agriculture that understand regional climate patterns and local soil idioms—not just imported agritech solutions from Silicon Valley. Imagine electoral AI tools that enhance citizen participation, not just police it. These are not fantasies. They are policy choices—and Africa must choose.
Encouragingly, countries like Nigeria and South Africa are developing homegrown AI frameworks. Nigeria’s 2023 Draft National AI Policy promotes responsible, inclusive AI, while South Africa’s Presidential Commission proposes principles rooted in human dignity. Yet, these frameworks remain aspirational. Nigeria’s policy, for instance, lacks legal teeth and hasn’t yet translated into enforceable regulation. The process excluded broad-based civic input and failed to define clear mechanisms for algorithmic accountability or impact assessments. South Africa’s Commission, meanwhile, has been criticized for being too industry-heavy and technocratic, sidelining voices from marginalized communities. As a result, these strategies—while forward-looking—risk being captured by elite interests and failing to address systemic digital inequalities.
Conclusion: Clicking “I Agree” on Our Own Oppression?
We are at a decisive inflection point. Africa can either become the training ground for imperial AI systems or the birthplace of a context-aware, ethical intelligence paradigm, one rooted in Ubuntu, communal accountability, and social justice. If colonialism once arrived with the cross and the crown, today’s digital order arrives with code and consent screens. But the effect is strikingly similar: the quiet disempowerment of African people under the guise of progress.
In April 2025, over 1,000 delegates; including heads of state, ministers, technologists, and civil society leaders, convened in Kigali, Rwanda, for the inaugural Global AI Summit on Africa. Under the theme “AI and Africa’s Demographic Dividend: Reimagining Economic Opportunities for Africa’s Workforce,” the summit aimed to chart a path for Africa’s active participation in the global AI ecosystem. A significant outcome was the signing of the Africa Declaration on Artificial Intelligence by representatives from 52 African countries, committing to responsible, inclusive, and sustainable AI development aligned with the continent’s priorities.
Rwandan President Paul Kagame emphasized the need for Africa to harness AI’s potential, stating, “Africa must not be just a market for AI but an active player. To achieve this, we need to build strong infrastructure, train our people, and promote inclusive AI.” Similarly, Nigerian Minister of Communications, Innovation and Digital Economy, Bosun Tijani, highlighted the importance of defining national AI goals to foster effective collaboration without compromising sovereignty.
These commitments are commendable. However, the challenge lies in translating declarations into actionable policies and frameworks. Many African nations still grapple with inadequate digital infrastructure, limited AI expertise, and insufficient regulatory mechanisms to govern AI effectively. Without addressing these gaps, there’s a risk that well-intentioned initiatives may not yield the desired outcomes, leaving citizens vulnerable to the very issues these efforts aim to mitigate.
This moment calls for more than compliance, it demands resistance. A new social contract must be forged not in Silicon Valley, but in the minds and hands of African citizens. We must center the teacher in Kano, the farmer in Eldoret, the market woman in Ibadan, not just the policymaker in Abuja or the data scientist in Cape Town. Civic literacy, participatory governance, and tech justice must form the backbone of our digital future.
If we continue clicking “I Agree” without scrutiny, we risk waking up in a society governed not by our elected leaders, but by opaque recommendation engines, i.e systems we did not build, cannot interrogate, and may never fully understand. Dear compatriots, this action would not simply be a betrayal of democracy; it would be the digitization of colonization itself.
Habeeb Adewale is a Postgraduate Student of Research and Public Policy at University of Lagos and Social Development expert with over 8 years of experience in managing projects that leverage technology for social goods.
LinkedIn: linkedin.com/in/habeeb-adewale-tajudeen-685464132
Twitter (X): @Habeeb_Adewale_
