Facial recognition technology eliminates the need for touching a device to enable access
Walk into a shopping mall in Durban, Cape Town or Johannesburg today and your face has probably been scanned, logged and analysed — all without your knowledge or consent. Facial recognition technology is spreading rapidly across South Africa, slipping into public spaces under the banner of “safety” and “convenience”. In truth, it is eroding one of our most basic rights — the right to be anonymous in public.
The spread of surveillance goes far beyond shopping centres. Gated estates demand facial scans at their entrances, quietly storing the biometric details of domestic and other workers and visitors with zero transparency and no consent. Security companies assure us this is “for our protection” but they never answer the hard questions: “Who owns the data? How long is it kept? Who can access it — and for what purpose?”
Even in our personal lives, we contribute to this growing system. Phones that unlock with our faces or fingerprints generate digital identity trails we rarely consider. These trails can be — and often are — accessed by corporations, law enforcement and hackers. A tool marketed as convenience can quickly become an instrument of surveillance.
To grasp the full scale, it helps to understand how the technology works. Facial recognition systems capture facial features, translate them into a digital algorithm and compare them against vast databases. This makes it possible to identify anyone caught on camera in seconds, often without their knowledge.
But recognition is only the beginning. Newer applications claim to predict behaviour, skills and even personality traits from facial measurements. Companies already use this data to infer gender, race and sexual orientation. These “facial judgments” echo the racial profiling that scarred South Africa under apartheid.
Back then, people were forced into rigid racial categories — white, native, coloured or Asian — determined not by choice but by degrading, pseudo-scientific tests. The infamous “pencil test”, in which a pencil was placed in a person’s hair to see if it would fall out or stay lodged, epitomised the absurd cruelty of a system obsessed with classification. These labels dictated every aspect of life: where one could live, which schools one’s children could attend, which jobs were open, even which hospital ward one would be treated in.
Surveillance and control held this racial order in place. Pass laws required black South Africans to carry identity documents at all times, granting or denying permission to move through white-controlled areas. Police could stop, question and detain anyone suspected of being “out of place”. The result was not only restriction of movement but also a society built on fear — where simply existing in the wrong space could be criminalised.
Today, facial recognition risks reviving this same logic, only cloaked in the sleek language of “safety” and “efficiency”. Where the apartheid state relied on passbooks and racial categories, biometric systems now rely on cameras and algorithms. Instead of a police officer demanding papers, a camera silently scans your face. Instead of raids, databases quietly compile patterns of behaviour, travel and association.
The technology is new, but the principle is chillingly familiar — reducing people to data points, stripping them of privacy and making surveillance a condition of belonging.
Defenders of this technology often insist that “if you’ve got nothing to hide, you’ve got nothing to fear”. But that misses the point. Constant, indiscriminate monitoring reshapes society itself. It normalises being watched, treats everyone as a potential suspect and slowly erodes the freedoms we take for granted. Once entrenched, such systems are nearly impossible to dismantle.
South African law is not silent on this. The Protection of Personal Information Act (Popia) classifies facial images as “special personal information”, which may only be processed with explicit consent, a legal obligation or a narrowly defined public duty. Popia also requires that people be notified when their data is collected. In practice, this almost never happens. Shopping centres, estates and private security firms operate these systems with little regard for the law and virtually no oversight.
Section 9 of Popia goes further, demanding that personal information be collected in a way that is adequate, relevant and not excessive. A blanket, real-time dragnet of every face entering a mall or estate is the very definition of excessive.
The tragedy is not that we lack laws but that enforcement has been toothless. Most South Africans are unaware that these practices are unlawful. And surveillance societies rarely announce themselves — they creep in quietly, through small shifts in what people are willing to accept.
Cape Town already offers a glimpse of where this leads. The city has poured resources into surveillance infrastructure, deploying roughly 2 000 closed-circuit television cameras across the metropolitan area. The technology is now unavoidable in daily life, yet remains largely unregulated. Its patterns of monitoring and profiling echo those of apartheid, reminding us that history has ways of repeating itself in new forms.
The uncomfortable truth is that South Africa is sleepwalking into a world where being watched becomes the default. Unless we begin asking the hard questions — and demanding accountability — we could wake up too late, in a country where privacy is no longer a right, but only a memory.
Sõzarn Barday is a writer and attorney based in South Africa and has a particular interest in human rights within the Middle East. The views expressed are her own.