Government Risk Profiling: The Hidden Danger of Discrimination

Margriet Vermeer · 2026-05-07

When the government uses data to predict who might commit a crime or commit fraud, it sounds like a smart, modern approach. But there is a dark side to this practice. A recent study from the Universiteit van Amsterdam warns that without principled measures, the risk of discrimination and harm remains too great. ### What Is Risk Profiling? Risk profiling is when agencies use algorithms and data analysis to flag individuals or groups as potential threats. Think of it like a credit score, but for law enforcement, tax audits, or immigration checks. The idea is to allocate resources more efficiently. But here is the problem: these systems often reflect the biases that already exist in society. For example, if historical arrest data shows more arrests in certain neighborhoods, the algorithm will flag those neighborhoods more often. That creates a feedback loop. More police presence leads to more arrests, which confirms the algorithm's prediction. And before you know it, entire communities are treated as suspects. ### Why This Matters Discrimination is not just a moral issue. It is a practical one. When people feel targeted unfairly, they lose trust in the government. That makes them less likely to cooperate with law enforcement or follow the rules. It also wastes taxpayer money. A system that flags innocent people over and over is not efficient. It is just expensive. The researchers from the Universiteit van Amsterdam point out that the problem is not the technology itself. It is the lack of principled measures. Without clear rules and oversight, these systems can cause real harm. They can turn a tool meant to keep us safe into a weapon of bias. ### What Needs to Change So, what do principled measures look like? Here is a quick list: - Transparency: Agencies must explain how their algorithms work and what data they use. - Accountability: There should be independent oversight to catch and correct errors. - Fairness: Algorithms must be tested for bias before they are deployed. - Human review: Automated decisions should always have a human in the loop. - Data privacy: Citizens should know what data is collected and how long it is kept. These are not radical ideas. They are basic safeguards. Without them, risk profiling can easily become a tool for discrimination. ### The Real-World Impact Imagine you are a small business owner in a city with heavy surveillance. Your shop is in a neighborhood that the algorithm flags as high risk. Suddenly, you get audited every year. Your customers get stopped by police on the way in. Your insurance rates go up. None of this is based on anything you actually did. It is just the algorithm's guess. That is not hypothetical. It is happening right now in cities across the United States. And it disproportionately affects communities of color and low-income neighborhoods. The harm is real. It damages lives, businesses, and entire communities. ### A Better Way Forward The answer is not to abandon technology. It is to use it wisely. That means involving communities in the design process. It means testing for bias before deployment. And it means being honest about the limits of what these systems can do. We have the tools to build fairer systems. We just need the will to use them. The research from the Universiteit van Amsterdam is a wake-up call. It reminds us that without principled measures, the risk of discrimination and harm remains too great. Let us not ignore that warning. Let us act on it.