Algorithmic surveillance turns traces of behaviour into judgments about who a person is and what they may do next.
Surveillance once suggested a person watching from a fixed position. Digital systems make observation continuous, distributed, and difficult to see. A search, purchase, pause, location signal, friendship, or viewing choice can become an input into a system that classifies a person without announcing that a judgment is being made.
Algorithmic surveillance is surveillance performed through computational rules. The system collects or receives data, combines it with other data, and produces an inference or decision. It can recommend, rank, target, approve, restrict, price, manage, or flag. The crucial shift is from recording what happened to estimating what a person means, wants, deserves, or will do.
That estimate does not need to be correct to become powerful. A profile can shape which opportunities appear, which messages reach someone, and which version of reality is repeatedly presented to them. The system’s judgment becomes part of the environment in which the person acts. Prediction turns into influence, and influence creates more data for the next prediction.
This is why algorithmic surveillance exceeds the familiar question of privacy. The issue is not only whether information about a person has been collected. It is whether an opaque system can convert that information into a category that governs the person’s options, while making the category difficult to inspect, contest, or escape. Group profiling makes the problem sharper because the affected person may never be individually identified. A population can be sorted and acted upon before anyone knows exactly who has been judged.
The practical response is cognitive and institutional sovereignty. People need to understand that convenience often carries an observing system inside it. Institutions need limits on collection, explanation of consequential inferences, meaningful avenues for challenge, and accountability for the outcomes produced by automated systems. Human judgment cannot remain a ceremonial signature added after the machine has already decided.
Source: Kosta, E. (2020), “Agency and algorithmic state surveillance,” Regulation & Governance, on data aggregation, prediction, pre-emption, and the challenge to legal agency. The definition and recommendation-system framing are also grounded in McHendry’s Key Concepts in Surveillance Studies (2019). Citation tier: peer-reviewed scholarship, supported by an academic open textbook.