AI Data Collection in Nigeria Raises Questions About Privacy and Control

AI’s value increasingly depends on how much it can learn from people. That raises a practical question for Nigeria: as more people use AI services for work, education and communication, who controls the information those tools collect? Access may widen, but the terms of participation still matter.

Data collection is becoming harder to see

Searches, social media activity and conversations with AI tools can leave behind records of what people care about and how they behave. A single interaction may say little. Accumulated across services, those records can reveal interests, routines and preferences that companies may use to develop products, personalize services or generate revenue.

The concern is not that every company collecting data has harmful intentions. It’s that commercial systems can reward collecting more information than users expect, while leaving them with limited insight into what happens to it. Privacy policies and user agreements rarely make that imbalance easy to assess.

A September 2024 investigation by the United States Federal Trade Commission examined nine major social media and video streaming companies, including Meta, YouTube, TikTok and Amazon-owned Twitch. The FTC reported that the companies collected extensive information about users and, in some cases, people who did not use their services. That information included personal details, interests, online activity and data from other sources.

The findings also raised concerns about long retention periods, inadequate privacy protections and the use of personal information in automated systems, including AI. The core issue was an information gap: companies could build detailed records, while individuals might not know how their information was being used or how to limit particular uses.

AI increases the stakes because data can help train models, refine recommendations and support automated decisions. More information can make these systems more capable, but it also expands the need for clear limits on what companies collect and what they do with it.

Access needs to come with informed choice

AI can offer practical benefits where access to resources and professional services is uneven. Students may use it to support learning; small businesses can apply it to daily operations; journalists can use it in research and production. Tools that work well with Nigerian languages and reflect local experience could also make digital services more accessible.

But access alone doesn’t make inclusion meaningful. People need to understand the conditions under which they use a service and retain rights over the information they provide. Many services present those conditions in lengthy policies that users may accept without grasping the consequences.

AI products also raise specific questions about prompts, uploaded files and other interactions. Depending on the product, users may want to know whether that material is retained, how long it remains available and whether it can be used to improve a model. Practices vary by company and service, so users need direct explanations rather than broad assurances.

For each service, the useful questions are concrete: What information does it collect? Why does it need that information? How long does it keep it? Could it be used beyond the task a person asked the tool to perform? Clear answers would make it easier to weigh an AI tool’s benefits against the information it requires.

Corporate data can create surveillance capacity

Private data collection and authoritarianism are not the same thing. Authoritarianism involves power exercised without adequate accountability and in ways that threaten rights and freedoms. Still, large stores of corporate data can make surveillance easier by enabling profiling, targeted persuasion and predictions about individual behavior.

The risk grows when data can be accessed or shared without adequate protections, when automated decisions carry serious consequences without meaningful oversight, or when people have little ability to challenge how their information is used. Companies, governments and other actors may be able to use the same capabilities in different ways.

Market concentration adds a practical constraint. When a small number of companies control widely used platforms, AI services and large pools of data, switching providers may be difficult. Leaving a platform can mean losing access to customers, professional networks, communities or essential services. In that setting, a nominal choice may offer little real control.

That doesn’t mean every AI system is a surveillance tool or that every technology company aims to restrict freedom. It does mean the infrastructure people rely on deserves scrutiny: tools built for commercial purposes can acquire uses beyond the ones their users first had in mind.

Nigeria’s framework depends on enforcement and awareness

Nigeria’s Data Protection Act 2023 established the Nigeria Data Protection Commission and set out a framework for lawful, fair and accountable processing of personal information. Having a legal framework matters, but protection also depends on effective implementation, public awareness and the ability to hold organizations responsible for their practices.

As AI use grows, people need accessible information about their rights and organizations need to explain what they collect and how they protect it. These obligations become more pressing when an AI product draws on information users may not realize they have supplied or retained.

Representation is another part of the issue. If AI systems don’t reflect Nigerian languages, cultures and social realities, their benefits may reach communities unevenly. Responsible governance therefore has to address both personal data protection and whether different communities are represented in the systems being built.

UNESCO’s Recommendation on the Ethics of Artificial Intelligence offers principles that can inform this work: privacy, data protection, transparency, accountability and fairness throughout an AI system’s lifecycle. Those principles point toward a standard that considers both what a system can do and how it affects the people whose information helps make it work.

Control is part of digital inclusion

Companies can develop useful technologies with data, but that doesn’t require treating privacy as an acceptable cost. They can limit unnecessary collection, explain their policies in clear language, offer meaningful choices and protect information from unauthorized access. Regulators, in turn, need to ensure that the existing framework works as AI products change.

Digital inclusion should account for more than how many people can access AI. It should also consider whether people can use these tools safely, understand the terms and exercise meaningful control over their information. For Nigeria, that means pairing wider access with clearer practices and stronger accountability.

The questions facing users and institutions are straightforward: What can AI companies learn from people’s data, who decides how those insights are used, and can individuals challenge those decisions? The answers will help determine whether AI expands people’s choices or quietly makes control over personal information harder to exercise.