Last month, I sat across from a venture capitalist at a dinner in Singapore who described his latest fund’s thesis as “digital public infrastructure for emerging markets.” He was animated, genuine, clearly convinced he was backing something meaningful. The portfolio included biometric ID systems for West Africa, cashless payment rails for Southeast Asia, and an AI-driven credit scoring platform designed for rural India. Every product required citizens to hand over intimate personal data — fingerprints, iris scans, location history, spending patterns — to systems architected thousands of miles away in San Francisco and Palo Alto.
I asked who owned the data. He looked at me like I’d asked about the weather.
“The platform,” he said. “Obviously.”
The architecture of extraction has a new name
There’s a pattern in how technology gets deployed across the Global South that we’ve been reluctant to name clearly. It goes like this: A government in Africa, South Asia, or Latin America identifies a legitimate governance problem (financial exclusion, lack of identity documents, poor healthcare access). Silicon Valley firms, often backed by development finance from the World Bank or USAID, offer a digital solution. That solution requires mass data collection. The data flows to servers controlled by private companies, often incorporated in Delaware or the Cayman Islands. The host country gets a dashboard. The company gets a dataset worth billions.
We call this development. We call it financial inclusion. We call it modernization. What it actually is: a new form of structural compliance, where populations with the least geopolitical power are enrolled into systems of surveillance they didn’t design, can’t audit, and frequently can’t opt out of.
The scale is staggering. India’s Aadhaar system has enrolled over 1.3 billion people in biometric identification. Kenya’s Huduma Namba, modeled partly on Aadhaar, faced enormous public resistance before being struck down by the courts over privacy concerns, only for the government to push forward with amended versions. In Uganda, SIM card registration tied to national ID has effectively made anonymous communication impossible. Each of these systems was framed as a gift: modernity delivered to the doorstep.
I’ve spent enough time reporting on tech ecosystems across Asia and Europe to recognize the difference between infrastructure built with communities and infrastructure imposed on them. The former looks messy, contested, slow. The latter looks clean, scalable, and backed by a pitch deck. The latter is what gets funded.
Who builds the rails controls the traffic
Here’s the structural reality. When a government in the Global South adopts digital identity or fintech infrastructure built by foreign firms, it creates a dependency relationship that mirrors colonial-era resource extraction with uncomfortable precision. The raw material is no longer rubber or copper. It’s behavioral data: how people move, what they buy, who they talk to, when they’re sick, where they worship.
A report by the Transnational Institute documents how digital ID programs across Africa are frequently conditioned on World Bank loans, meaning countries adopt these systems under financial pressure rather than democratic consensus. The report notes that in many cases, the technical architecture is designed abroad, maintained by foreign contractors, and governed by terms of service that override local data protection laws (where they exist at all).
This matters enormously. When you control the identity layer of a society, you control who can access government services, open a bank account, receive a subsidy, or prove they exist in the eyes of the state. These are not market efficiencies. These are instruments of political control, and they’re being handed to private companies whose fiduciary duty runs to shareholders, not citizens.
I think about the way we talk about the European tech ecosystem, where there’s constant hand-wringing about sovereignty, about keeping data within EU borders, about ensuring GDPR compliance. That conversation, however imperfect, at least acknowledges that data governance is a question of power. The equivalent conversation barely exists for most of the Global South, because the power differential is so vast that the question isn’t even raised. It’s assumed away.
The compliance loop
What makes this dynamic particularly insidious is the feedback mechanism. Once a population is enrolled in a digital identity system tied to financial services, the system doesn’t just observe. It disciplines.
Consider how credit scoring works in markets where formal credit histories barely exist. Companies like Tala, Branch, and dozens of smaller operators use phone data (call logs, SMS patterns, app usage, contact lists) to generate credit scores for people who’ve never had a bank account. The pitch sounds empowering: access to capital for the unbanked. The reality is that these scores create a behavioral regime where people learn, quickly, that their digital behavior is being monitored and evaluated. Miss a payment and your score drops. Uninstall the app and you lose access to financial services. Your phone becomes a leash.
A 2023 study published in Big Data & Society found that digital lending platforms in East Africa frequently accessed users’ entire contact lists and would contact borrowers’ family and friends to pressure repayment. The researchers described this as “networked shaming,” a form of social discipline that weaponizes community ties. The platforms defended it as a necessary innovation for “high-risk markets.”
This is the compliance loop: offer a service that people genuinely need (credit, identity documents, healthcare), make it contingent on data extraction, then use the data to enforce behavioral norms defined by the platform. The citizen becomes a user. The user becomes a data point. The data point becomes a risk score. And the risk score determines whether you eat.
The development-industrial complex
I want to be precise about who benefits from this arrangement, because it isn’t a conspiracy. It’s a market.
The development finance ecosystem (the World Bank, the International Finance Corporation, bilateral aid agencies) has spent the last decade pivoting toward “digital development.” This means funding flows increasingly toward technology platforms rather than physical infrastructure. The logic is seductive: digital solutions scale faster, cost less per unit, and generate data that can be used to optimize further interventions. From the perspective of a development economist in Washington, this looks like progress.
From the perspective of a farmer in Madhya Pradesh who must authenticate via iris scan to receive her food subsidy, and whose authentication fails 10% of the time due to worn fingerprints or server outages, it looks like a new bureaucracy with a silicon veneer. Researchers have documented cases where Aadhaar authentication failures led to denied food rations in India, contributing to documented starvation deaths. The system was eventually reformed in specific ways, but the fundamental architecture (your body as your password, your biometrics as your access key) remains.
Meanwhile, the companies building these systems are valued in the billions. The venture capital firms backing them report to limited partners who expect returns. The consulting firms advising governments on “digital transformation” bill by the hour. There is an entire industry that profits from the surveillance of the Global South, and it operates under the banner of inclusion.
I’m not writing from a position of purity here. I run a media company. I operate within the same attention economy that monetizes user data, that optimizes for engagement, that treats people as metrics. The difference is one of degree and consent. Most users of European or American tech platforms at least nominally agree to terms of service, can switch providers, and live under legal regimes that offer some recourse. For a subsistence farmer whose government has mandated biometric enrollment as a precondition for receiving grain, the concept of “consent” is theater.
Sovereignty as the contested ground
The most important fights in global technology policy over the next decade will be about data sovereignty, and they’ll be fought most fiercely in places where Silicon Valley has already embedded itself deepest.
There are promising signs of resistance. Rwanda has built its own digital identity infrastructure with explicit sovereignty provisions. Brazil’s Pix system, a government-run instant payment platform, was designed to prevent private monopolies over payment data. India itself has begun to push back against foreign control of its data, even as its domestic surveillance apparatus grows more powerful.
But resistance requires capacity: legal, technical, and political. Many of the countries most aggressively targeted by digital development programs are precisely those with the least capacity to negotiate terms, audit code, or enforce data localization. The asymmetry is the feature.
I keep returning to a question that I think the European tech ecosystem, for all its navel-gazing, has at least begun to grapple with: who gets to decide what data governance looks like? In Europe, the answer (imperfectly, unevenly) is democratic institutions. In the Global South, the answer is increasingly a venture-backed startup with a mandate to grow.
What honest language sounds like
Language matters in this conversation because the euphemisms do real work. “Financial inclusion” obscures the fact that inclusion into a privately controlled financial system is also a form of capture. “Digital public infrastructure” obscures the fact that the infrastructure is frequently neither public nor governed as such. “Emerging markets” obscures the fact that these markets were deliberately kept from emerging by the same geopolitical powers now selling them software.
If we’re honest about what’s happening, we’d say: populations in the Global South are being enrolled into surveillance architectures designed to extract behavioral data and enforce compliance with market norms defined elsewhere. The enrollment is incentivized through access to basic services. The data flows upward. The accountability doesn’t flow at all.
This is a technology story, but it’s really a power story. And power stories don’t resolve through better product design or more thoughtful pitch decks. They resolve through politics: through movements, through regulation, through the slow, unglamorous work of building institutions that can say no.
The venture capitalist I had dinner with is a good person. His fund will likely generate strong returns. The platforms he backs will probably reach millions of people who currently lack access to formal financial services. And those millions of people will have been quietly enclosed into a system that watches everything they do, scores their behavior, and reports to shareholders in another hemisphere.
We should stop calling that development. We should call it what it is.
Feature image by cottonbro studio on Pexels
