Why Hiding Is Getting Harder — Even for the Most Vulnerable
In an era where a glance can turn into a data point and a walk down the street leaves a digital trail, a new warning is emerging from the intersection of technology and civil liberties. A powerful opinion piece published by The Washington Post argues that the accelerating capabilities of artificial intelligence — particularly facial recognition and identity-linking tools — are rapidly eroding the ability to remain anonymous. For millions of people who depend on low-profile living for physical safety or personal freedom, this technological shift is not just a privacy concern; it is an existential threat.
Written by James Harkin, director of the Center for…, the column frames AI-driven identification as a quiet revolution in surveillance that turns everyday environments into persistent observation networks. Harkin’s central thesis is that “hiding” — a deliberate, often life-preserving strategy — is becoming dramatically more difficult. The piece, which is also featured by the AI-focused newsletter Superintelligent, examines how vulnerable populations such as activists, migrants, abuse survivors, and whistleblowers are losing the protective cloak of anonymity just when they need it most.
“Hiding is becoming more difficult as AI systems improve at matching faces, tracking identities, and linking data from multiple sources.”
The column emphasizes that facial recognition is only one node in a vastly larger surveillance ecosystem. Social media scrapers, public cameras, data brokers, and automated profiling tools now stitch together fragments of a person’s identity — a photo uploaded years ago, a license plate scan, a location check-in — into a cohesive, searchable dossier. What makes the current moment so concerning, Harkin warns, is not the existence of any single technology but their convergence and real-time interoperability. A person fleeing an abusive partner, for instance, can be pinpointed by combining smart doorbell footage with geotagged social media content and readily available data broker records, often without a warrant or meaningful oversight.
The consequences are starkest for those without institutional power. Activists organizing against authoritarian regimes have long relied on pseudonymity and careful separation of their online and offline identities. Migrants escaping violence may need to move through cities undetected to avoid apprehension or retaliation. Survivors of domestic violence often go to extraordinary lengths to scrub their digital footprints. For all these groups, Harkin suggests, the margin of safety is shrinking as AI-driven identification becomes cheaper, faster, and more pervasive. A single facial match can unravel years of rebuilding a life away from danger.
Beyond individual tragedies, the opinion piece draws attention to a fundamental policy deficit. Legal frameworks around the world were designed for a time when anonymity was the natural default in public spaces. Today, that default has been inverted. Biometric surveillance is expanding across airports, retail stores, and even places of worship, often with minimal public debate. Harkin points out that while regulators scramble to draft rules for generative AI or autonomous vehicles, the quiet normalization of permanent digital identifiability proceeds with far less scrutiny. Civil liberties organizations such as the Electronic Frontier Foundation have long documented the risks of unregulated facial recognition, and the United Nations human rights office has repeatedly called for safeguards around privacy-invasive biometrics.
Harkin’s Washington Post column is careful to note that its argument is not anti-technology; rather, it is an urgent call to bring democratic accountability to tools that can permanently reshape the boundary between public and private life. As AI systems edge closer to real-time, continuous identification across crowded spaces, the editorial contends that society must decide whether preserving the right to disappear — even temporarily — is a value worth protecting. Without deliberate legal action, the piece concludes, the cost may be measured in human lives that could have been shielded if only the technology had been matched by ethics.
