Meta just rolled out a new AI tool that’s turning heads and raising alarms. It tracks employees’ every click, mouse movement, and keystroke. The goal? To train AI systems that can take over routine computer tasks. Sounds futuristic, right? But here’s the twist: this tool isn’t just watching U.S. workers. It’s scooping up data from European employees too. That’s a big problem under strict EU privacy laws.

What Is Meta Tracking and Why?

The tool, dubbed the Model Capability Initiative (MCI), is designed to observe exactly how Meta’s U.S. employees interact with their computers. It watches everything — from the websites they visit, to code changes, to the tiny clicks that navigate dropdown menus. Why? To build AI agents that could someday perform these tasks automatically.

But the data collection goes deeper than expected. MCI pulls from over 200 apps and websites. It even logs clipboard contents and computer sleep-wake cycles. It’s like creating a digital fingerprint of how each employee works.

Meta says this helps them train AI models with real-life examples. The company insists it’s focusing on how people use computers, not the content on their screens. But the tool captures emails and chats between U.S. staff and their international colleagues. That means European employee data ends up in the mix, unintentionally or not.

Why EU Regulators Are Raising Red Flags

The European Union’s General Data Protection Regulation (GDPR) has some of the toughest rules on data privacy worldwide. It demands companies have a clear legal reason for collecting personal data. It also requires full transparency about what data is gathered and why.

Here’s the catch: MCI collects data originally meant for work communication. Using it to train AI models changes the purpose. Privacy experts say this violates GDPR’s “purpose limitation” principle. Basically, you can’t use personal work chats for something else without fresh consent.

Meta has told regulators in Ireland that capturing EU employee data “does not fall within the primary purpose” of the tool. But privacy advocates are skeptical. Even incidental data capture could trigger serious legal trouble.

• Data collected includes emails and instant messages with U.S. staff
• Employee mouse clicks, keystrokes, and navigation patterns are logged
• The tool connects to existing security software, accessing unencrypted info
• Massive data volumes can build detailed behavioral profiles of workers

Legal experts warn this could amount to unauthorized employee surveillance. That’s a major privacy risk and a potential breach of EU law.

Employee Backlash and Data Security Concerns

Inside Meta, the reaction has been intense. Workers call MCI an “Employee Data Extraction Factory.” They feel they are training the AI systems that might replace them. Some complain the tool devours so much bandwidth it burns through their entire home internet quota in days.

An internal analysis revealed MCI logs highly sensitive data. That includes source code edits, clipboard content, URLs visited, and even computer sleep cycles. Shockingly, this data is stored unencrypted. That raises big questions about security and misuse.

One Meta employee described the system as creating a “complete behavioral model” of how a knowledge worker performs their job. Not just clicking dropdowns — but knowing exactly what to select, where to paste, and what to do next.

Meta’s spokesperson insisted these claims are inaccurate but declined to answer detailed questions. Meanwhile, some internal posts discussing these findings mysteriously disappeared.

What’s Next for Meta and AI Surveillance?

This controversy is far from over. Meta faces rising pressure from EU regulators to explain how it handles employee data. Privacy watchdogs call for investigations. The stakes are high — this case could set a precedent for AI training and employee privacy worldwide.

Meta’s broader mission to automate work through AI is accelerating. But this tool’s backlash highlights a crucial challenge: balancing innovation with privacy rights. Can Meta build smarter AI without crossing legal and ethical lines?

For employees, the fear is real. Are they simply training the machines that will replace them? For the tech world, this sparks urgent questions about trust, transparency, and the future of work.

One thing is clear: AI surveillance at work isn’t just a tech issue. It’s a privacy battle that’s just heating up. And everyone—from regulators to employees—will be watching closely.