Meta has quietly announced it will begin monitoring how its employees work—tracking their keystrokes, mouse clicks, and digital work patterns—to generate training data for the company's artificial intelligence systems.

The decision, reported by BBC Technology, represents one of the most invasive workplace surveillance programs implemented by a major technology company to date. While many organizations monitor employee productivity, Meta's approach goes further by converting routine work activity into fuel for machine learning models.

What Meta Is Actually Tracking

According to the announcement, Meta will collect granular data about how employees interact with their computers during work hours. This includes keystroke patterns, mouse movements, click frequencies, and potentially the sequence of applications and tools workers use throughout their day.

The company frames this as an opportunity to improve AI systems by learning from real human work behavior. By observing how experienced employees navigate complex tasks, Meta's algorithms could theoretically learn more natural and efficient ways to automate similar processes.

But what Meta calls "learning from human expertise," privacy advocates might describe differently: comprehensive digital surveillance of workers who likely have limited ability to opt out without career consequences.

The Consent Problem

The most troubling aspect isn't necessarily what Meta is collecting—it's the power dynamic that makes meaningful consent nearly impossible.

When your employer announces a new monitoring system, can you really say no? For Meta employees, particularly those on work visas or in competitive roles, refusing to participate could carry professional risks the company doesn't need to articulate.

Meta has not publicly detailed whether employees can decline this monitoring, what happens to those who object, or how long the collected data will be retained. These aren't minor implementation details—they're fundamental questions about workplace rights in the age of AI.

The company also hasn't clarified whether this data will be anonymized, how it will be secured, or what prevents it from being used for performance evaluation despite any stated intentions.

A Precedent Other Companies Will Watch

Meta's decision matters beyond its own workforce because of the precedent it establishes. When one of the world's largest technology companies normalizes this level of employee monitoring, it provides cover for countless other organizations to follow suit.

Workplace surveillance has expanded dramatically over the past decade, accelerated by remote work during the pandemic. Employee monitoring software already tracks time spent in applications, websites visited, and even takes periodic screenshots. Meta's approach simply adds AI training as a new justification for data collection that was already happening in many workplaces.

The difference is scale and purpose. Meta isn't just monitoring for productivity—it's harvesting behavioral data to build commercial AI products that will compete in a multi-billion dollar market.

What This Means for Workers Everywhere

If this practice spreads, millions of workers could find their daily computer interactions becoming training data for AI systems—often without meaningful compensation or control over how that data is used.

Consider what your keystroke patterns reveal: how fast you type, how often you pause, when you make corrections, which tasks slow you down. Mouse movements show hesitation, confidence, efficiency. Together, these data points create an intimate portrait of how you think and work.

This information has obvious value for AI training. It also has value for performance algorithms, worker replacement strategies, and productivity optimization systems that might not serve employee interests.

The Regulatory Vacuum

Meta's announcement highlights how far workplace surveillance has outpaced legal protections. In the United States, employers have broad latitude to monitor workers, especially on company-owned devices. European regulations like GDPR provide somewhat stronger protections, but enforcement remains inconsistent.

No comprehensive framework governs whether companies can use employee work product as AI training data. No clear standards exist for consent in workplace monitoring programs. No regulations specifically address the unique privacy implications of harvesting human behavioral data to train machine learning systems.

Meta is operating in this vacuum, and the absence of guardrails is precisely what makes the announcement possible.

The Company's Perspective

To be fair, Meta would likely argue that learning from employee workflows is no different than any organization studying how its teams work to improve processes. The company has built its business on collecting user data to improve products, and applying similar logic to internal operations isn't necessarily hypocritical.

AI systems do need training data, and observing skilled humans performing complex tasks is a legitimate way to gather it. If Meta can build better AI tools by understanding how its own employees work, that could benefit both the company and potentially its users.

The question isn't whether this data has value—it clearly does. The question is who controls it, who benefits from it, and whether employees have genuine agency in deciding whether to participate.

What Employees Should Know

If you work at Meta, you should understand what's being collected and ask direct questions about your rights. Can you opt out? Will this data be used for performance reviews despite assurances otherwise? How long will it be retained? Who has access to it?

If you work elsewhere, watch whether your employer follows Meta's lead. The normalization of AI-driven workplace surveillance is happening now, and it will be easier to push back early than after these systems become entrenched.

The Bigger Picture

Meta's keystroke monitoring program is a symptom of a larger transformation in how companies view employees in the age of AI. Workers aren't just performing tasks—they're generating valuable behavioral data that can be extracted, analyzed, and converted into commercial AI products.

This shift reframes the employment relationship in troubling ways. You're not just trading labor for compensation anymore. You're also, perhaps unknowingly, contributing training data for systems that might eventually automate your role or optimize your work in ways that benefit shareholders more than workers.

The technology industry has spent years collecting data from users to train AI systems. Now it's turning that same extractive logic inward, treating employees as another data source to be harvested.

Whether this becomes standard practice across corporate America, or whether workers and regulators push back successfully, will shape workplace privacy for decades to come. Meta has made its choice. The question is whether employees, lawmakers, and the public will accept it.