AI is turning office monitoring into minute-by-minute surveillance

AI-enabled monitoring systems now process huge volumes of worker data to produce minute-by-minute accounts of employee activity and productivity, a shift the Guardian examined in a video segment published on its Science channel on October 15, 2026.

The segment, hosted by Guardian science editor Ian Sample, features the paper's UK technology editor Robert Booth, who explains which sectors are adopting AI workplace surveillance, why employers are deploying it, and how the practice could prove counterproductive.

The story matters because workplace monitoring is not new, but its scale and granularity are changing. That shift raises questions about productivity measurement, employee trust, and the boundaries of legitimate management oversight — questions that policy makers and employers are only beginning to address.

How did we get to 'bossware'?

Workers have always been monitored by their bosses, as the Guardian frames the historical arc. The first wave relied on output targets: employers measured what workers produced and held them to quotas or goals.

The second wave moved monitoring onto the computer. Software installed on employees' machines — labelled 'bossware' — tracked digital activity during working hours. This category of tool became a fixture in many offices, particularly after remote work expanded the distance between managers and the people they manage.

The third wave is the current one. AI transforms the capabilities of these systems. Where earlier tools logged discrete events, AI-driven versions process huge volumes of data and provide what the Guardian describes as minute-by-minute accounts of activity and productivity.

That is a categorical change, not an incremental one. Continuous, granular measurement differs in kind from periodic reporting, because it removes the intervals in which work was previously assessed in summary form.

Which sectors are adopting AI monitoring?

Booth, in conversation with Sample, explains which sectors are turning to AI-based workplace surveillance and the motivations behind the adoption. The Guardian's segment positions this as a broad trend rather than a niche experiment: the technology is spreading across multiple industries, and employers are deploying it for reasons tied to measurement and management of productivity.

The video format does not enumerate every sector by name in its summary, but the framing is explicit: adoption is wide enough that the Guardian's UK technology editor treats it as a structural shift in how work is observed, not as an isolated practice.

Why are employers doing it?

The stated logic, as laid out in the segment, is productivity. AI monitoring promises employers a finer-grained picture of what workers do during the day — activity tracked continuously rather than reconstructed from output alone.

For managers, the appeal is visibility. Minute-by-minute data promises to replace inference with measurement. Where an output target told a boss only whether a result arrived, AI surveillance claims to show the process: when work happened, how much, and at what pace.

Whether that promise holds is precisely where the Guardian's segment turns skeptical.

How could it backfire?

Booth explains how AI workplace surveillance could end up being counterproductive, and that warning anchors the segment's analytical core.

The counterproductivity argument follows a familiar mechanism: monitoring that pressures workers can degrade the very behavior it aims to improve. Systems that rank employees by continuous activity metrics invite gaming — workers optimizing for the metric rather than the underlying work.

There is also the trust dimension. Surveillance shifts the employment relationship toward suspicion. Workers who know they are watched minute by minute may comply visibly while disengaging substantively, a response that no activity dashboard will capture.

The Guardian does not present this as a marginal concern. The editorial decision to foreground counterproductivity — rather than treat surveillance as a neutral management tool — signals that the risks are central to the story, not a caveat appended to it.

What does the shift mean for the workplace?

The stakes extend beyond individual offices. Continuous AI monitoring redefines what employers can know about workers, and that capability outpaces the norms and rules governing its use. The gap between technical capability and regulatory constraint is where the significant conflicts will emerge.

The trajectory the Guardian describes — output targets, then bossware, then AI-driven minute-by-minute tracking — shows each wave of technology expanding the observable surface of work. Each expansion has met pushback, and the current one is large enough that its resolution will shape employment practice across sectors.

For workers, the practical question is what counts as activity in an AI system's accounting, and what that classification does to people whose work is not easily rendered as measurable events. For employers, the question Booth raises is whether the granular data buys productivity or erodes it.

Where does this go next?

The Guardian's segment closes on the tension it opened with: AI has given workplace monitoring unprecedented reach, and that reach may work against the productivity gains employers are chasing. As adoption spreads across sectors, the debate moves from whether employers can monitor at this granularity to whether doing so pays — a question the counterproductivity warning suggests will be answered in many workplaces the hard way.

Marcus Bennett covers AI policy and the future of work for AI In Context.