Watched, Measured, and Discarded: How Workplace Surveillance Became the New Whip of American Labor
The Eye That Never Blinks
In a fulfillment center outside Memphis, a warehouse associate pauses for thirty seconds to stretch her back after lifting her four-hundredth package of the shift. Within moments, a productivity dashboard flags her as underperforming. A supervisor receives an alert. The associate, who has worked the same floor for three years, is issued a verbal warning before lunch.
This is not a hypothetical. Variations of this scene play out thousands of times each day across the American economy. Employers — from logistics giants to insurance firms to fast-food franchises — have invested heavily in what the technology industry euphemistically calls "workforce management solutions." Strip away the branding, and what remains is a surveillance apparatus of extraordinary reach: keystroke loggers, biometric time clocks, GPS trackers embedded in delivery vehicles, AI-powered cameras that parse facial expressions for signs of distraction, and algorithmic scoring systems that reduce a human being's labor to a decimal point on a performance index.
The scale of this infrastructure is not incidental. It is the product of deliberate choices — choices made by executives and investors, ratified by a legal system that has consistently prioritized managerial authority over worker dignity, and obscured by the language of efficiency, security, and optimization.
A Two-Tiered System of Accountability
Perhaps the most revealing feature of modern workplace surveillance is not what it captures, but whom it targets.
Studies examining monitoring practices across industries consistently find that intensive, real-time surveillance is concentrated among lower-wage workers: warehouse associates, call center representatives, delivery drivers, home health aides, retail clerks, and remote customer service employees. These are workers who already possess the least bargaining power, the fewest legal protections, and the most precarious employment arrangements.
Senior managers, executives, and knowledge workers in professional-class occupations operate under an almost entirely different set of expectations. Their outputs are measured quarterly, if at all. Their movements are not tracked by GPS. Their keystrokes are not logged. Their bathroom breaks are not timed. The asymmetry is not accidental — it reflects a deeply embedded assumption about whose time is valuable, whose judgment can be trusted, and whose body is, ultimately, an instrument of production rather than a sovereign self.
This is class discipline rendered in code. The surveillance infrastructure does not simply measure productivity; it enforces a social hierarchy, signaling to lower-wage workers that their autonomy is conditional, their privacy negotiable, and their humanity secondary to the throughput metrics that govern their days.
Algorithmic Management and the Erosion of Judgment
Beyond the psychological weight of constant observation, workplace surveillance systems carry a more structural danger: they transfer decision-making authority from human supervisors — who can exercise contextual judgment, recognize individual circumstances, and respond to appeals — to algorithmic systems that cannot.
When a warehouse algorithm generates a termination recommendation based on "time off task" data, it does not know that the worker paused to assist a colleague who had dropped a heavy load. It does not register that the same worker has a spotless safety record over four years. It processes inputs and produces outputs, and in many workplaces, those outputs carry the force of managerial decisions without the accountability that human management at least nominally implies.
Amazon's much-documented history of algorithmic terminations — in which workers were dismissed through automated processes with minimal human review — represents an extreme but instructive example. The company has faced sustained criticism and some legislative scrutiny for these practices, yet the underlying model has proliferated across the logistics, retail, and gig economy sectors. When the algorithm decides, there is no one to appeal to, no supervisor to humanize the interaction, and no mechanism through which a worker's full context can be considered.
This is not efficiency. It is the systematic removal of accountability from the exercise of power.
The Health Costs No Dashboard Measures
The consequences of pervasive surveillance extend well beyond disciplinary outcomes. A growing body of occupational health research documents the physiological and psychological toll of working under continuous monitoring. Elevated cortisol levels, increased rates of anxiety and depression, higher incidence of musculoskeletal injury — these are not abstractions. They are the predictable results of forcing human beings to operate at machine-calibrated pace without the recovery time that biological organisms require.
At Amazon warehouses, injury rates have been documented at roughly twice the industry average for serious incidents, a pattern that safety researchers have linked directly to the productivity quotas enforced through algorithmic monitoring. In call centers, workers subject to real-time performance dashboards report significantly higher rates of burnout and emotional exhaustion than those in less intensively monitored environments.
The costs of these health outcomes are not borne by the corporations that design the surveillance systems. They are externalized onto workers, their families, and the public health infrastructure that absorbs the downstream consequences. The surveillance tax, in other words, is paid not by shareholders — who benefit from the productivity extracted — but by the workers whose bodies and minds absorb the pressure.
Legislative Terrain and the Organizing Imperative
The legal framework governing workplace surveillance in the United States remains remarkably permissive. Federal law imposes few meaningful restrictions on employer monitoring of workplace communications, electronic activity, or physical movement. Several states — California, New York, and Connecticut among them — have enacted disclosure requirements compelling employers to inform workers that monitoring is occurring, but disclosure is a far cry from limitation. Knowing you are being watched does not diminish the harm of being watched.
A handful of jurisdictions have moved more ambitiously. New York City's 2022 law regulating the use of automated employment decision tools — requiring bias audits and candidate notification — represents one of the more substantive interventions at the local level. California's Privacy Rights Act extends some protections to employee data. But these measures remain islands in a largely unregulated sea.
The most consequential responses have come not from legislatures but from organized workers. Warehouse workers in Staten Island who formed the Amazon Labor Union did so in part out of frustration with algorithmically enforced quotas. Call center workers at multiple firms have bargained surveillance limitations into their contracts. Delivery drivers at UPS, represented by the Teamsters, have negotiated language restricting the use of telematics data in disciplinary proceedings.
These victories are instructive. They demonstrate that the surveillance apparatus, however technically sophisticated, is not immutable. It is a product of power relations — and power relations can be renegotiated when workers organize with sufficient clarity and collective will.
What Justice Demands
Dr. Ambedkar understood that formal legal equality means little when structural power remains concentrated in the hands of those who own the instruments of economic life. The surveillance systems that now govern the working hours of millions of Americans are precisely such instruments — tools through which employers extract maximum output while minimizing worker agency, dignity, and recourse.
Justice in this domain requires more than disclosure requirements. It requires enforceable limits on real-time algorithmic discipline, mandatory human review before adverse employment actions, worker access to the data collected about them, and genuine collective bargaining rights over the introduction and scope of monitoring technologies.
Above all, it requires that we name what these systems are: not neutral tools of management science, but mechanisms of class control — designed to serve capital, calibrated to discipline labor, and sustained by a legal and political order that has historically treated the interests of employers as synonymous with the public good.
The workers flagged by dashboards, terminated by algorithms, and injured by quotas deserve better than that equation. Building the political and organizational power to change it is not a peripheral concern of the justice movement. It is central to it.
