Sixty thousand AI-powered eyes will soon watch Hong Kong’s streets, and the end of anonymity in public may arrive much sooner than you think.
Story Snapshot
- Hong Kong will expand its surveillance network from 4,000 to 60,000 AI-enabled cameras by 2028.
- Real-time facial recognition is set to become operational by the end of 2025, with a pilot already running in police patrol cars.
- Officials claim the move is vital for national security and crime prevention, but critics warn of privacy erosion and unchecked police power.
- The initiative mirrors mainland China’s surveillance model and raises urgent questions about oversight and civil liberties.
Hong Kong’s Surveillance Expansion: The End of Public Anonymity?
Hong Kong’s surveillance ambitions are no longer whispers—they are policy. The city’s police-led “SmartView” program will ramp up the number of CCTV cameras from about 4,000 today to a staggering 60,000 by 2028, most equipped with facial recognition powered by artificial intelligence. The first wave will appear in police patrol cars this July, and by the end of 2025, select cameras will scan faces in real time, matching passersby against vast, police-controlled databases. Authorities insist this is a logical step to strengthen national security and curb crime—yet absent from the official narrative is any real discussion of how this data will be controlled, who will hold the police accountable, or whether Hong Kong’s famed public spaces can ever feel the same again.
This is not Hong Kong’s first dance with digital eyes. The city’s surveillance system began its rapid growth after the tumultuous 2019 pro-democracy protests, when masked crowds clashed with riot police and the government’s grip on dissent wavered. The 2020 National Security Law—imposed from Beijing—ushered in a new era of monitoring, granting authorities sweeping powers to surveil, detain, and prosecute. Since then, over 10,000 people have been arrested in the name of stability. The SmartView program is the logical next phase: an infrastructure that promises “order” while quietly redrawing the boundaries of public life.
Who Holds the Power, and Who Watches the Watchers?
The cast of decision-makers is clear, and so is the hierarchy of power. The Hong Kong Police Force, led by Security Chief Chris Tang, will run the SmartView program—implementing, operating, and expanding it with little independent oversight. The Office of the Privacy Commissioner for Personal Data (PCPD) is officially tasked with safeguarding privacy, but sources confirm the commissioner was not consulted before this expansion. Civil society groups, legal scholars, and ordinary citizens are left with questions and unease, especially as the city’s government borrows playbooks from mainland China’s “Sharp Eyes” and “Skynet” surveillance projects. Hong Kong’s public is both the target and the test case, especially those in high-traffic and protest-prone neighborhoods.
Hong Kong to install surveillance cameras with AI facial recognition https://t.co/6sKFndwZwG
— Geo Political Updates (@GeoPolUpdates) October 5, 2025
Hong Kong’s lawmakers have been briefed, but the legislative process has become a formality; the police and government hold the real levers. The Privacy Commissioner’s recommendations—such as requiring privacy impact assessments and clear public signage—have been referenced, yet there is scant evidence these safeguards will meaningfully curb police discretion. The public’s direct influence is minimal, but international scrutiny and pressure could still shape the city’s reputation as a global financial hub.
Security or Suppression? The Stakes of AI Surveillance
Proponents point to the stats: since its inception, the SmartView program has been credited with solving 400 cases and leading to 787 arrests. For authorities, the logic is simple—more cameras, faster identification, safer streets. But privacy advocates and legal scholars see a darker calculus. They warn that AI facial recognition is notoriously error-prone, risking wrongful arrests and amplifying government power without transparency or recourse. The lack of clear rules or independent oversight leaves the door wide open for abuse, especially in a city where political activism and dissent have already become dangerous pursuits.
Hong Kong’s dense cityscape, history of protest, and status as a regional crossroads magnify these risks. The normalization of mass surveillance could erode civil liberties, chill free expression, and foster a culture of self-censorship. Meanwhile, AI and surveillance technology vendors stand to profit, and regional governments are watching closely. Will Hong Kong’s experiment become a blueprint—or a cautionary tale?
The Road Ahead: Open Questions and Uncertain Futures
By October 2025, the rollout will intensify, with public announcements and media scrutiny mounting as real-time facial recognition draws nearer. The full 60,000-camera network is set to span housing estates, metro stations, roads, and public squares by 2028. Yet uncertainty lingers: officials have not confirmed exactly when all cameras will gain AI capabilities, and the details of how footage and biometric data will be stored or accessed remain opaque.
For Hong Kong’s residents, the implications are immediate and personal. The city’s transformation into a “smart” metropolis may bring a sense of safety to some, but for many, it signals the end of true anonymity in public. The debate over security versus freedom has rarely been starker, and the coming years will reveal whether the city’s embrace of AI surveillance is a shield—or a cage.
