The Quiet Violence of Digital ID: Who Gets Locked Out and Why

Digital identity systems are sold to us as a tidy upgrade. Efficiency. Security. Inclusion. Governments and global agencies frame them as the obvious next step—a way to pull the unbanked into the financial fold, smooth out public services, and shut down fraud. But if you look past the polished policy briefs and the pilot-program press releases, a messier truth takes shape. For a staggering number of people, the move from paper to pixels doesn’t open a door. It quietly bolts it shut.

I’ve spent years watching how identity infrastructure shapes who gets to claim their rights and who gets ignored. The conversation usually orbits around the tech: biometric accuracy, blockchain immutability, interoperability standards. What gets sidelined is the human architecture of vulnerability. When a state or a corporation designs a digital ID system, it bakes in a set of assumptions about what a “normal” user looks like. Those assumptions radiate outward, building barriers that are invisible to the designers but absolute for the people who crash into them.

The Assumption of a Fixed Identity

Digital ID rests on a quiet premise: that your identity attributes stay put. A name, a birth date, a fingerprint—the system treats them as constants. For many people, they’re anything but. Refugees fleeing conflict may never have had a birth certificate. Transgender and nonbinary individuals often carry documents that clash with their lived gender, and updating those records can mean navigating a bureaucratic maze that digital portals only make more rigid. Someone sleeping rough doesn’t have a fixed address, yet plenty of enrollment forms treat that field as mandatory. The system demands stability from lives defined by disruption.

Take stateless populations. The UN estimates at least 10 million people worldwide have no nationality. They lack the foundational documents—passports, national ID cards—that most digital ID schemes require as a starting point. A digital credential is supposed to solve the problem of missing paper, but if you need paper to get the digital credential, the promise collapses. The system doesn’t fill the gap; it digs it deeper, wrapping the old exclusion in a new layer of cryptographic certainty.

Biometric Barriers and Bodies That Don’t Fit the Mold

Biometrics—fingerprints, iris scans, facial recognition—get pitched as the great equalizer. No literacy required, no paperwork to lose. The reality is rougher. Fingerprint scanners routinely fail for manual laborers whose ridges have been sanded down by years of physical work. Iris scans stumble over certain eye conditions or post-cataract surgery eyes. And facial recognition? A 2019 National Institute of Standards and Technology study found that many commercial algorithms were up to 100 times more likely to misidentify Black and Asian faces than white ones. That’s not a rounding error. That’s a structural failure.

These aren’t fringe scenarios. They describe huge swaths of the global population. When a digital ID system leans on biometrics as the main gatekeeper, it systematically penalizes people whose bodies fall outside the algorithmic norm. The failure isn’t a bug. It’s a design choice that quietly marks certain bodies as standard and others as deviations.

Digital Literacy and the Access Gap

Even if you have the right documents and your body reads cleanly to the scanner, another hurdle sits waiting: the tech itself. Digital ID systems often assume a smartphone, a steady internet connection, and a baseline comfort with apps and portals. Older adults, people with disabilities, rural residents with spotty connectivity, and those with limited formal schooling can find themselves staring at a screen that offers no way in.

India’s Aadhaar system—the largest biometric digital ID program on the planet—has enrolled over a billion people and earned plenty of praise. But field studies keep uncovering stubborn problems. Fingerprint authentication failures at ration shops have turned away families who rely on subsidized food. Elderly beneficiaries, their fingerprints faded by age, get denied at the point of sale. The system was built to curb corruption. Instead, it sometimes manufactures a new kind of exclusion: the exclusion of bodies that can’t perform the required digital transaction.

Economic Exclusion Through Digital Gatekeeping

Digital ID is increasingly lashed to financial services. Mobile money, digital wallets, online banking—all of them demand verified digital identities. For people living in poverty, the cost of getting and keeping a digital ID can be a wall. A smartphone, a data plan, a reliable way to charge the device: none of these are universal. In sub-Saharan Africa, where mobile money has exploded, women are far less likely than men to own a mobile phone. That gap in device ownership translates directly into a gender gap in digital financial inclusion, layering new inequality on top of old.

Then there’s the problem of losing the credential. A paper ID can be folded, tucked away, carried. A digital ID that lives on a phone vanishes the moment the phone is stolen, smashed, or simply runs out of battery. For someone scraping by on the economic margins, replacing a device may be out of reach. The digital credential, sold as more secure, turns out to be more brittle in the context of poverty.

Surveillance and the Chilling Effect on Marginalized Groups

Digital ID systems don’t just verify who you are. They generate a trail. Every time you use one to access a service, cross a border, or make a payment, that interaction gets logged. For communities already under pressure—ethnic minorities, political dissidents, LGBTQ+ people in repressive environments—that traceability can be dangerous. A digital ID that ties your biometrics to your location history, your purchases, and your use of social services builds a surveillance machine that hostile state actors can weaponize or that data breaches can spill into the open.

In some countries, digital ID systems have been used to track and restrict the movements of refugees, penning them into camps or capping their access to work. The same technology that promises inclusion for some becomes a tool of containment for others. The design question isn’t whether the system can track people. It’s whether it will—and who gets to make that call.

Administrative Violence in Digital Form

Anthropologist Akhil Gupta gave us the term “administrative violence” to name how bureaucratic procedures harm people, often without intent, through neglect, arbitrariness, and structural indifference. Digital ID systems can amplify that violence by automating decisions that human officials used to make—officials who could, at least sometimes, exercise discretion. When an algorithm denies a benefit because a fingerprint doesn’t match, there’s no appeals officer to talk to. The system just says no.

This automated exclusion hits hardest for people who already live at the edges of state recognition. Someone with an irregular migration status. Someone whose name is spelled differently across their documents. Someone whose gender presentation doesn’t line up with their official records. These individuals once depended on the flexibility of human bureaucrats to navigate the cracks. Digital ID systems, with their rigid data fields and automated verification, strip that flexibility away. What’s left is a harder, less forgiving border around the category of “legible” citizens.

Designing for the Margins

The problems I’ve laid out aren’t inevitable. They grow from a design philosophy that starts with the average user and treats edge cases as exceptions to be mopped up later. A more principled approach would flip that: start with the most vulnerable users and build outward. Some call it “universal design” or “inclusive design,” but the label matters less than the practice.

What would a digital ID system look like if it were designed first for a stateless person, a transgender refugee, an elderly manual laborer with worn fingerprints, and a rural woman without a smartphone? It would probably offer multiple authentication pathways, not just biometrics. It would make room for self-asserted identity attributes alongside verified ones. It would work offline and on basic mobile phones, not just smartphones. It would bake in strong privacy protections that limit data collection and sharing by default. And it would include human-mediated fallback processes for when the technology inevitably stumbles.

Some of these features are starting to surface. Estonia’s digital ID infrastructure lets citizens control what data different service providers can see. The MOSIP (Modular Open Source Identity Platform) project, adopted in several countries, includes offline authentication. But these are the exceptions. The dominant models, pushed by development banks and technology vendors, still favor centralized biometric databases and smartphone-centric interfaces.

The Role of Civil Society and Community-Led Design

Meaningful change won’t come from technical tweaks alone. It demands a shift in who holds power during the design process. Right now, digital ID systems are overwhelmingly shaped by engineers and policymakers who sit at a long distance from the daily realities of excluded populations. Community-led design processes—where affected groups have genuine decision-making authority—can surface needs and constraints that outside experts miss entirely.

For instance, organizations working with unhoused populations have documented how digital ID systems that demand a permanent address effectively block access to shelters and services. In consultation with the people who navigate these systems every day, some municipalities have started issuing digital credentials that use a shelter or service center as a proxy address. Those solutions didn’t emerge from a technical specification document. They came from listening.

Civil society organizations also play a watchdog role that no design sprint can replace. In Kenya, the rollout of a biometric national ID system triggered legal challenges from groups worried about data protection and the exclusion of marginalized communities. The resulting court rulings forced the government to pause mandatory enrollment and address privacy safeguards. Without that organized pushback, the system would have gone live with its exclusionary features intact.

Policy Recommendations for Inclusive Digital ID

After years of watching these systems operate in the wild, I have a few concrete recommendations for policymakers and implementers:

• Mandate multi-modal authentication. No single biometric or credential should be the only gate. Systems should accept a range of proofs—community attestations, self-asserted attributes—with graduated levels of assurance depending on how sensitive the transaction is.
• Guarantee offline functionality. Digital ID systems must work without internet connectivity, using stored credentials on low-cost devices or printed QR codes that can be verified manually.
• Enshrine data minimization in law. Collect only the attributes necessary for a specific transaction. Prohibit the creation of centralized biometric databases that can be repurposed for surveillance.
• Fund independent oversight bodies. These bodies should include representatives from marginalized communities and have the power to audit systems, investigate complaints, and order changes.
• Require human fallback processes. Every automated decision must have a clear, accessible, and timely path for human review and appeal.
• Conduct exclusion impact assessments. Before deployment, systems should be rigorously tested with diverse populations to identify who will be left out and why.

These recommendations aren’t radical. They’re basic safeguards that any system handling fundamental rights ought to include. The fact that they’re often missing from major digital ID initiatives tells you how far the design conversation has drifted from a rights-based framework.

Frequently Asked Questions

Why are digital ID systems promoted as inclusive when they can exclude people?

Proponents focus on the potential to reach people who lack paper documentation, particularly in developing countries. The argument is that biometrics can identify someone without a birth certificate. That’s true in some cases, but it skips over the many ways digital systems create new barriers—biometric failure, cost, connectivity, and data privacy risks—that land hardest on the same vulnerable populations the systems claim to serve.

Can’t exclusion problems be fixed with better technology?

Some technical improvements can help: more accurate fingerprint scanners that work with worn ridges, facial recognition algorithms trained on genuinely diverse datasets. But technology alone can’t solve problems rooted in social inequality and political choices. A system that requires a smartphone will still shut out people who can’t afford one, no matter how elegant the software is. Addressing exclusion demands policy changes, not just engineering updates.

What can individuals do if they are excluded by a digital ID system?

First, document what happened. Write down the details, the time, the error message or denial you got. Seek out civil society organizations or legal aid groups that work on digital rights or social welfare access. In some countries, ombudsman offices or data protection authorities can investigate complaints. Collective action—joining with others who’ve faced similar exclusion—can also pressure authorities to create alternative access pathways.

Are there examples of countries doing digital ID well?

Estonia is often cited for its user-controlled data model and strong legal protections. Canada’s approach emphasizes provincial autonomy and privacy-by-design principles. Still, even these systems have gaps, particularly for Indigenous populations and people experiencing homelessness. No system is perfect, but the better ones share common features: legal limits on data use, multiple authentication options, and mechanisms for public accountability.

The conversation about digital ID needs to move past techno-optimism and stare hard at the distributional consequences of these systems. Who benefits? Who bears the costs? Who gets rendered invisible? These aren’t questions to tuck into the final paragraph of a project report. They’re the questions that should frame the whole endeavor. Until they do, digital ID will keep being a tool that works beautifully for those already inside the circle of recognition—and brutally for those on its edges.