So, your boss wants to install an AI on your computer to make sure you’re working. Not just clocking in, but really working. Every tab you open, every idle moment, every glance at a personal message—all fed into a machine that decides if you’re a “good” employee. Sound like a dystopian film plot? It’s not. It’s the rapidly emerging reality of workplace AI ethics, and it’s forcing a very uncomfortable conversation we can no longer afford to ignore.
This isn’t about blocking a few websites. This is about a fundamental shift in the power dynamic between employer and employee, where algorithms are becoming the new middle managers. And as we dive into tools designed for employee surveillance under the guise of productivity, the privacy boundaries we once took for granted are being redrawn in real-time, often without our consent.
Just What Are We Talking About With Workplace AI Ethics?
Let’s be clear. Workplace AI ethics isn’t some fuzzy, academic notion. It’s the set of rules—or, more accurately, the glaring lack of them—governing how companies use artificial intelligence to manage, monitor, and measure their staff. For years, AI in the office was about optimising supply chains or sorting customer emails. Benign, behind-the-scenes stuff. Now, it’s personal.
The technology has crept from the server room to your desktop, promising to unlock human potential but often delivering a digital panopticon instead. The ethical challenge isn’t the technology itself, but the choices we make about its deployment. Do we use AI to empower people, or to police them? Right now, the scales are tipping alarmingly towards the latter. The conversation is no longer about if AI will be used, but how it will be used, and who gets to write the rules.
The All-Seeing Eye: AI as the Ultimate Taskmaster
The new wave of employee surveillance tools goes far beyond counting keystrokes or logging active windows. We’re now in the age of attention tracking, a concept that sounds as creepy as it is. These systems aim to understand not just what you’re doing, but why you’re doing it. Are you researching a work topic on Reddit, or are you just doom-scrolling?
Think of it like having a personal trainer who, instead of helping you with your form, simply shouts every time you pause to catch your breath. It’s a system built on a foundation of distrust. It presumes that, left to your own devices, you will inevitably slack off. This constant monitoring has a corrosive effect on autonomy and creativity, turning work into a performance for an algorithmic audience. How can anyone do deep, meaningful work when a digital supervisor is perpetually peering over their shoulder?
More Data, More Problems
The core tension with these AI monitoring tools lies in the conflict between their stated goals and their actual function. They promise to boost productivity using sophisticated performance metrics, but in reality, they often just create anxiety.
– Flimsy Privacy Boundaries: Many tools claim to anonymise or redact personal information, but this is often a technical fig leaf. When screenshots of your screen are being uploaded to a third-party cloud, can you ever be truly certain your private messages, bank details, or sensitive work data are secure? What happens when—not if—that data is breached?
– The Illusion of Objective Metrics: AI-driven performance metrics offer a veneer of objectivity, but they are only as good as the data they are trained on. An algorithm can’t tell the difference between staring at a blank page while having a breakthrough idea and staring blankly because you’re bored. The result? A workforce optimised for looking busy, not for being effective.
This obsession with constant measurement creates a dangerous paradox. In the quest for an extra 5% of efficiency, are organisations willing to sacrifice 100% of their employees’ trust and psychological safety?
Case Study: Is Fomi Your Friend or a Foe in Disguise?
Let’s look at a prime example that perfectly encapsulates this dilemma: a macOS app called Fomi. As detailed in a recent WIRED article, Fomi positions itself as a sophisticated solution to distraction. Its creator, Zach Yang, notes that “defining what counts as ‘work’ versus ‘distraction’ is pretty nuanced,” and he’s not wrong. Fomi’s answer is to use OpenAI’s GPT-5 Mini model to analyse the content of your screen, deciding in real-time if you’re on task.
A Peek Under the Bonnet
On the surface, it’s clever. The app doesn’t just block Twitter; it tries to understand if you’re using it for market research or just watching cat videos. It provides gentle, colour-coded nudges to get you back on track. For a monthly subscription of around £6.50 ($8) after a trial, it promises to be the focus coach you never knew you needed.
But here’s the rub. To achieve this contextual understanding, Fomi takes screenshots of your activity and sends them to the cloud for processing. The developers say they redact personally identifiable information (PII) locally before uploading. Still, the app transfers roughly 500MB of your screen data to a remote server every single day.
The Privacy Conundrum
This is where the alarm bells should be deafening. Entrusting your screen’s contents to a cloud service, even a “redacted” version, is a massive leap of faith. It’s like posting your private journal online after blacking out a few names, hoping nobody can connect the dots.
The privacy policy might promise security, but the architecture itself is the problem. Why does this processing need to happen in the cloud at all, exposing users to potential data breaches and mission creep? As cited by WIRED, the convenience of cloud-based AI comes at a steep, and perhaps unacceptable, privacy cost. Fomi represents a critical juncture: a tool designed for individual productivity that perfectly illustrates the corporate surveillance risks on the horizon.
Finding the Tightrope: A Path Forward?
Is it possible to reap the benefits of AI without succumbing to a culture of surveillance? Perhaps, but it requires a radical commitment to transparency and trust that few organisations currently demonstrate. “Balancing” productivity and privacy is the wrong frame; one should not come at the expense of the other.
Instead of top-down monitoring, organisations could focus on using AI to remove friction and automate tedious tasks, freeing up employees to do the work that matters. The key is consent and control. Any tool that monitors an employee’s work should be fully transparent in its methods, and the employee should have ultimate control over their data. The goal should be augmentation, not automation of management.
The Future Is Watching
The rise of tools like Fomi isn’t a niche trend; it’s a preview of the future of work if we don’t get a handle on workplace AI ethics right now. This technology will only become more powerful, more integrated, and more capable of making judgments about our value as workers.
The challenge is not to stop innovation, but to steer it in a direction that respects human dignity and autonomy. We need to ask harder questions. Who benefits from this data? What are the long-term consequences of a work culture built on algorithmic oversight? And are we willing to trade our privacy for a phantom promise of productivity?
The time for passive acceptance is over. It’s time for employees, leaders, and technologists to demand a better bargain. What are your red lines when it comes to AI in the workplace?
