In December 2025, Amazon rolled out an internal dashboard that tracks employees' individual badge swipes, how long they stay in the building, and which building they used — refreshed daily, visible to managers and HR, and rolled up over an eight-week window feeding directly into performance reviews. Meta, around the same time, deployed AI-usage monitoring dashboards that flag employees' non-work activity during business hours and tie the results to project reviews.
Neither company is an outlier. They're just the biggest names in a shift that's already reached most American workplaces: as of 2026, 78% of companies use employee monitoring software, and 80% of remote workers are monitored in some form. This isn't the return-to-office debate. It's what's actually waiting at the desk once you show up.
What "Monitoring" Actually Means in 2026
The badge-swipe dashboards at Amazon and Meta are the visible tip of a much larger stack. Employee monitoring software — sold under brand names like Teramind, Hubstaff, and ActivTrak — routinely includes keystroke logging, random or continuous screenshot capture, mouse-movement and idle-time tracking, full website and application usage logs sorted into "productive" and "unproductive" categories, and in the most invasive tier, webcam-based facial recognition and emotion detection layered on top of email and chat content scanning.
Some of these tools go further still, incorporating biometric signals — stress indicators, even body temperature — into an AI-generated "productivity score" that can feed directly into performance reviews, raises, and terminations without an employee ever seeing the underlying data that produced the number.
The Numbers Behind the Boom
This isn't a fringe trend. Seventy-one percent of employees globally now report experiencing some form of digital monitoring at work, and 61% are subject to AI-powered surveillance systems specifically. The employee monitoring software market is projected to hit $4.5 billion, and over 500 documented workplace monitoring disputes and complaints have piled up as the tools have spread from call centers and warehouses into white-collar office work.
The stated justification is almost always productivity and security. The actual research on outcomes tells a more complicated story: 59% of monitored workers report damaged trust in their employer, 42% say they plan to quit within a year, and 45% report high stress specifically attributable to the surveillance itself. Worse for the productivity argument companies actually make: monitoring appears to reduce genuine output, pushing workers toward what researchers call "productivity theater" — jiggling a mouse, keeping a green status light on, optimizing for what the software measures rather than for the work itself.
The Legal Gray Zone
U.S. federal law offers remarkably little specific protection against any of this. A patchwork of state rules is starting to fill the gap — New York now requires employers to disclose electronic monitoring to employees, and California has proposed restrictions specifically targeting automated discipline decisions and workplace facial recognition — but there's no comprehensive federal framework governing what employers can track, how long they can keep it, or what they're allowed to do with an AI-generated score derived from it.
There's one federal angle with real teeth, though it's still underused: the Consumer Financial Protection Bureau has warned that AI-driven workplace surveillance tools which generate scores used in employment decisions may trigger obligations under the Fair Credit Reporting Act — the same law that governs credit reports — because a "productivity score" that affects your job can legally function like a background-check report. FCRA violations carry statutory damages of $100 to $1,000 per affected employee, which is a meaningfully larger stick than most workplace privacy complaints currently have available. Outside the U.S., enforcement has already shown up with real consequences: Amazon was fined roughly €32 million in France in 2024 specifically over "excessively intrusive" employee monitoring practices, under GDPR.
What You Can Actually Do
- Ask HR directly what's being monitored, and get the answer in writing. In states like New York, employers are legally required to disclose this — use that requirement even if you're not sure it applies to you.
- Read your monitoring software's actual data flow, not just the marketing page. "Workforce analytics" and "keystroke logging" are sometimes the exact same product with different branding aimed at employees versus IT buyers.
- Know your FCRA angle. If a monitoring tool generates a score used in a hiring, promotion, or termination decision, ask whether it's being treated as a consumer report under FCRA — most employers haven't thought this through, and asking the question alone sometimes changes how a tool gets used.
- Use hardware-level countermeasures for what software policy won't cover — a physical webcam cover or a hardware microphone blocker gives you certainty that software settings can't quietly override.
- Document the human cost, not just the policy. The trust and stress data above is exactly the kind of evidence that's moved state legislatures to act — individual accounts of burnout and anxiety, aggregated, are what turns "everyone complains about this" into an actual bill.
The Other Side
Employers point to real, legitimate reasons for at least some of this: security teams need to detect data exfiltration and insider threats, companies have a genuine interest in verifying remote employees are actually working the hours they're paid for, and badge data specifically has uses — building capacity planning, emergency evacuation accountability — that have nothing to do with surveillance at all. Some monitoring vendors argue their tools are explicitly designed to protect employees too, flagging burnout risk from overwork patterns rather than only policing slack time.
Critics counter that the productivity research cuts against the core business case even on its own terms — if monitoring measurably increases stress, damages trust, and pushes workers toward gaming the metrics rather than doing better work, then the primary beneficiary of most of these tools isn't the company's bottom line, it's the vendor selling the dashboard. The security argument is the strongest one employers have; the productivity argument is increasingly hard to make with the data now available, and it's telling that the tools rarely get pitched to employees as something that might also serve their interests until after the rollout is already underway.
Sources referenced: State of Surveillance, CurrentWare, Insightful, the Consumer Financial Protection Bureau, and reporting on Amazon's and Meta's internal monitoring dashboards.
Related gear
If webcam and mic-based monitoring is the part you can't stand, our USB-C Microphone Blocker gives you a hardware kill switch software can't override. The Ghost Protocol field guide covers know-your-rights basics that apply just as much at your desk as anywhere else.
0 comments