You might think Meta would simply just be laying off thousands of workers and end it there. Wrong!

It is asking its workers to teach artificial intelligence how to do their jobs, down to their mouse movements.

On April 21, 2026, It was reported that Meta would begin installing tracking software on the computers of its U.S.-based employees.

The system captures mouse movements, button clicks, keystrokes, and occasional screenshots during the workday.

The data feeds directly into AI training pipelines, helping Meta build agents capable of performing white-collar computer tasks on their own.

In an internal memo seen by Reuters and belonging to the Meta Superintelligence Labs team, staff were told they could "do their part to help just by doing their daily work."

That tells you everything you need to know.

The Slow Unraveling of a Workforce

To understand what is happening at Meta right now, you need the full picture.

This is not a company in crisis. Meta made $60 billion in profit in 2025 and is projecting $100 billion in U.S. social media ad revenues in 2026.

It is spending between $115 and $135 billion on AI infrastructure this year alone. It acquired a 49% stake in data-labelling firm Scale AI for $14.3 billion, and Scale's former CEO, Alexandr Wang now runs Meta Superintelligence Labs.

Despite this financial dominance, Meta is preparing to cut approximately 8,000 jobs in May 2026, representing about 10% of its global workforce. For context, these are not its first cuts.

READ: 8,000 Jobs Gone, More Coming: Meta Platforms Is Trading Workers for AI

In January, around 1,500 Reality Labs employees lost their jobs. In March, another 700 were let go across at least five divisions.

Since 2022, Mark Zuckerberg has eliminated roughly 25,000 positions. Executives have privately noted that future layoffs could be adjusted depending on how fast Meta's AI capabilities advance.

This simply means the more effectively the AI learns, the fewer humans are needed.

That is the loop: employees generate data; the data trains AI; the AI replaces the employees.

Your Workflow Is the Product

For Meta's official position, a company spokesperson provided this quote: "If we're building agents to help people complete everyday tasks using computers, our models need real examples of how people actually use them—things like mouse movements, clicking buttons, and navigating dropdown menus."

Meta has also stated that safeguards are in place to filter sensitive content, that data is encrypted and deleted after training, and that the collection is restricted to specific work applications and websites.

However, it is important to note that this data is not simply improving a productivity tool. It is being used to build autonomous AI agents capable of executing white-collar tasks independently.

Traditional job titles are being retired. In their place are roles like "AI builder," "AI pod lead," and "AI org lead."

About 1,000 employees have already been rebrandedand the restructuring is companywide.

A Generational Problem Dressed in Tech Language

If you are in your twenties or early thirties right now, you are entering or already inside a job market that is quietly being compromised at its foundations.

Across the tech sector alone, over 73,000 jobs have been lost in 2026 so far, according to Layoffs.fyi.

Amazon has cut roughly 30,000 corporate roles. Block eliminated nearly 40% of its workforce in February.

This extends far beyond tech. CBS News, Disney, and others have followed.

The Goldman Sachs analyst Jacqueline Arthur has pointed to emotional intelligence, judgment, and trust-building as the skills least likely to be displaced.

These are not skills you build through a coding bootcamp. They are built through years of working with other people, exactly the kind of experience that requires a job market willing to give you the chance.

What This Actually Means

There is a word for what Meta is doing to its employees, and it is not "efficiency." It is extraction.

Workers are being asked to hand over the intimate, practised knowledge of how they navigate their daily tasks, knowledge built through years of experience, to train systems specifically designed to make them redundant.

The only honest way to put this is that Meta's employees are not contributors to an AI initiative. They are its raw material.

For a generation already grappling with housing costs, unstable contracts, and the hollowing out of entry-level roles, this should be a big warning sign.