Human Rights in the Age of AI and Predictive Policing
Introduction
There has been a flux of rapid development in the field of technology. These changes have altered various segments of human society, and human rights are no exception. The rise of technology has consistently raised concerns about various aspects of rights, including privacy, freedom, and work, among others. One of the most significant impacts on these rights has been made by AI and Predictive Policing. This article will examine the current state of human rights in light of these significant developments.
The Context
Both AI and predictive policing rely on the collection and analysis of a massive amount of data. Predictive policing aims to help predict potential future crimes by either identifying the place where there is a risk of a crime being committed or the person who is likely to commit a crime. And AI enables computers and machines to stimulate human learning, comprehension, problem solving, decision making, creativity and autonomy.
Data, whether in collection or bias, is what worries human rights activists the most. This is because predictive systems treat data as neutral, even though it is produced by historically unequal policing and surveillance practices. As a result, existing social and institutional biases risk being reproduced and legitimised under the appearance of technological objectivity.
The Intent
The concept of human rights is a highly debated topic. Some argue that it is a Western imposition, while others advocate for separate tribal human rights. Rather than entering that debate, this article adopts a functional baseline: human rights are rights people have by virtue of being human, not because a state grants them. Their primary purpose is to place limits on the exercise of power, even when that power is exercised in the name of efficiency, security, or governance.
This also rejects the notion that freedom alone is sufficient without access to essential necessities such as food, shelter, health, and education. Reflecting this, the UN has classified it into two broad categories:
Civil & Political: Right to life, liberty, freedom from torture/slavery, expression, assembly, religion, fair trial.
Economic, Social & Cultural: Right to work, education, adequate living standards (food, housing, health), social security.
The Impact
AI and predictive policing systems are trained on vast amounts of historical data. The problem is that this data is not neutral. It reflects past policing practices, social prejudices, and unequal enforcement. As a result, the outputs generated by these systems often replicate the same biases present in the data on which they were trained.
This directly affects human dignity. For instance, predictive policing tools have disproportionately flagged Black men as potential criminals, not because of individual conduct but because of patterns embedded in historical records. Similarly, facial recognition systems have wrongly identified individuals due to poor or skewed datasets, leading to serious consequences based solely on automated outputs. Acting on such results without adequate human verification turns technological error into legal harm.
Supporters of predictive policing argue that AI removes human bias from policing. At first glance, this seems reasonable. However, this claim weakens when the system is trained on older data, collected at a time when discriminatory practices were more openly tolerated and less likely to face consequences. Instead of eliminating bias, the system risks preserving it in a less visible, more authoritative form.
Concerns also arise from the vague and expansive nature of data collection practices. When individuals are monitored, profiled, or categorised without clear legal limits, questions of liberty and dignity are inevitably triggered. The issue becomes even more severe when AI is deployed in automated warfare systems, where technology lowers the threshold for violence and makes large-scale harm easier to execute with reduced human accountability.
That said, the impact of AI on human rights is not entirely negative. When used responsibly, AI can expand access to education for children with limited resources, potentially advancing equality rather than undermining it. In legal systems burdened by delay, AI can assist by organising information and speeding up processes, provided it supports human decision-making instead of replacing it.
Similarly, while automation is often criticised for threatening jobs, it does not eliminate the right to work. In many cases, it reshapes the nature of work, allowing individuals to move away from stagnant roles and acquire skills for more suitable employment. The challenge lies not in the existence of AI, but in how and where it is used.
Conclusion
There is little doubt that AI and predictive policing pose serious challenges to human rights. Concerns surrounding large-scale data collection, surveillance, and the loss of privacy are unlikely to disappear anytime soon. These risks are structural and require constant legal scrutiny. At the same time, not all concerns stand on the same footing. Issues such as biased data and discriminatory outcomes, while real, are increasingly being acknowledged and addressed through improved training models, enhanced datasets, and clearer guidelines on how parameters are applied.
This distinction matters. If AI systems are trained responsibly and deployed within well-defined limits, they have the potential to reduce inequality rather than deepen it. In education, access to technology can help bridge resource gaps. In governance and legal systems, AI can assist overburdened institutions by organising information and supporting faster decision-making, so long as final authority remains with human actors.
The challenge, therefore, is not whether AI should exist within human rights frameworks, but how it should be used. A purely defensive approach that focuses only on preventing harm risks missing the broader opportunity. Instead of viewing AI solely as a threat, it is time to consider how it can be shaped to actively advance human rights objectives. The task ahead is not to reject technology, but to shape it so that it operates within the limits of human rights and actively contributes to their protection and promotion.
