Digital Dominance and Neo-Imperialism

In Abuja, the Nigeria AI Collective, under the Federal Ministry of Communications, Innovation and Digital Economy, hosted the BUILD Conference from October 21 to 23, 2026, themed “Building Africa’s Shared Compute and AI Infrastructure”. Policymakers, students, researchers, journalists, innovators and investors gathered to discuss how Africa can build, share and govern the computing power its AI future depends on. It also reminded us how Africa can break the chain of digital imperialism.

When Kwame Nkrumah published Neo-Colonialism: The Last Stage of Imperialism, in 1965, the parades for independence had barely ended. New flags were flying, new anthems were being sung, but Ghana’s founding president warned that empire had not died. It had simply changed its accounting.

If Nkrumah were alive today, he would recognise the method instantly. Only now it lives in the cloud.

Classical imperialism was brutally direct: redirect local resources to the metropole, stifle domestic industry, lock colonies into unequal trade, and draw arbitrary borders. The Berlin Conference of 1884 did it with rulers and maps. Today it is done with terms of service, cloud subscriptions and training data.

Consider the infrastructure that runs the modern world. Three American companies — Amazon, Microsoft and Google — control over 65 per cent of global cloud capacity, according to industry estimates. Africa, with 18 per cent of the world’s population, controls less than one per cent. That gap tells you who owns the foundational infrastructure of the digital economy, and who merely rents it.

News organizations across the continent increasingly rely on global tech platforms (social media and search engines) to distribute content, which subjects them to external algorithms and rule-setting.

This new face of neo-imperialism is not an abstraction. It is an extractive business model that threatens economic self-determination and is dismantling one of Africa’s most vital democratic pillars: its independent press.

The assault has come in three waves.

The first wave was distribution capture. Foreign tech conglomerates monopolised digital advertising, capturing over 80 per cent of local ad spend in key African markets, according to the Reuters Institute. Newsrooms that had survived military decrees and fuel scarcity suddenly found themselves dependent on algorithms in Menlo Park for their audience. When those platforms deprioritised news to maximise screen time, the floor collapsed.

As CEO of the Newspaper Proprietors’ Association of Nigeria, I have sat with publishers who cut 40 per cent of their newsrooms in the last two years not because readers left, but because the platform that delivered readers changed its mind. The audience is still there. The business model is not.

The second wave has been faster and more extractive. Generative AI has turned distribution capture into data colonialism.

Decades of African journalism — investigations produced under military regimes, during fraught elections and public health crises, often at great personal risk — are being scrapped without licence, credit or compensation to train large language models in California.

Those models then return our work in a form that kills the original: the zero-click summary. Why visit a Nigerian newspaper’s website when a chatbot synthesises its exclusive investigation into a single paragraph, offering no traffic, no attribution and no revenue to the newsroom that funded it?

Worse, because African content constitutes under two per cent of global AI training data, according to several AI governance audits, these summaries routinely default to crude Western stereotypes. When AI fails to comprehend Africa, it invents Africa. It is erasure by statistics.

Then comes the third wave, which should alarm everyone. Call it Stealth AI.

This is no longer a chatbot hallucinating. It is an industrial factory of synthetic fakes: artificial reporters, deepfake audio of political candidates on election eve, autonomous bot swarms manufacturing consensus on WhatsApp and Facebook. Threat researchers recently exposed international firms that invented a French-Ghanaian investigative journalist wholesale, name, byline, face, to plant coordinated smear campaigns in West African media.

More than 60 per cent of African journalists now say they struggle to distinguish genuine sources from synthetic fakes, according to surveys by the Centre for Journalism Innovation and Research. That confusion is the point. It’s good to note that the The Federal Competition and Consumer Protection Commission (the Commission, FCCPC, has opened formal inquiries into this practice, but the scraping continues daily.

But Stealth AI is only possible because of a deeper economic surrender. Africa is home to 1.5 billion people, 60 per cent under 25, speaking over 2,000 languages. Every day it generates billions of valuable data points — mobile money receipts, agricultural telemetry, Nollywood scripts, voice notes in Amharic.

This is the raw material of artificial intelligence. Yet where is it processed? Northern Virginia. Where is it stored? Dublin and Singapore. Who owns and monetises the resulting models? Companies headquartered in San Francisco, Seattle and Hangzhou.

We export raw data for free across platforms we do not own and subsea cables we do not control, only to import it back as an expensive finished product: a chatbot that stumbles over Hausa, a credit scoring model blind to informal markets, an educational tool unversed in Achebe.

This is why the conversation in Abuja must move from declaration to dominance. Compute — the combination of reliable energy, data centre capacity, fibre and specialised processors — is the railway and seaport of this century. Digital sovereignty is impossible without energy sovereignty. Whoever controls compute and power controls commerce.

What should that look like in practice?

First, we need a fair compensation framework. Just as Australia and Canada forced tech platforms to bargain with publishers, Africa needs a continental AI and data bargaining code. If your model was trained on our archives, you must licence it and pay for it. When a news organisation invests money to create content, it must get a fair share from its consumption.

And this cannot be glossed over: the press is an essential national and civic infrastructure. “Were it left to me” says Thomas Jefferson, a one time American President, “to decide whether we should have a government without newspapers or newspapers without a government, I should not hesitate a moment to prefer the latter”. At all times and through all ages, this national infrastructure must be jealously protected and defended.

Second, a sovereign storage mandate. All public data, government archives and data generated with public funds must be stored on cloud infrastructure physically located in Africa and subject to African law. No sovereignty without location.

Third, a concrete compute target. The African Union, through the African Continental Free Trade Area, should set and fund a goal: five per cent of global data centre capacity on African soil by 2030, powered by reliable energy. Ambitious, but not impossible — the continent already attracts over $2 billion annually in data centre investment; it now needs policy to direct and scale it.

The question is whether Africa will be a tenant or a landlord in it.

Adediran is Chief Executive Officer of the Newspaper Proprietors’ Association of Nigeria (NPAN).