At Meta, Profit and Surveillance Deepen Worker Unrest

A Profitable Meta Faces a Restive Work Force

Meta is heading into a fresh round of layoffs even as its profits soar, deepening a sense of disillusionment inside a company that has spent years pitching artificial intelligence as its future.

Beginning May 20, the company is expected to cut about 10 percent of its work force, or roughly 8,000 jobs, according to reports, while also leaving about 6,000 open roles unfilled. At the same time, employees in the United States and Britain are pushing back against an internal software program that tracks workplace activity on company laptops, including keystrokes, mouse movements and clicks, and some screen context, in an effort to improve Meta’s A.I. systems.

Together, the two developments have become a powerful symbol of a widening tension across the tech industry: companies are racing to build more capable A.I. tools while asking fewer workers to do more, under closer scrutiny, with less say over how their labor is used.

Inside Meta, employees have described morale as exceptionally low, the product of repeated cuts, tighter budgets and an “efficiency” drive that has reshaped the company since its earlier hiring boom.

The strains are especially striking because they come at a moment of extraordinary financial strength. In the first quarter of 2026, Meta reported $56.3 billion in revenue and $26.8 billion in net income, sharply up from $42.3 billion and $16.6 billion in the same period a year earlier. It also raised its planned capital expenditures for 2026 to between $125 billion and $145 billion, underscoring the scale of its bet on A.I. infrastructure.

That contrast — record profits alongside job cuts and employee monitoring — has become difficult for many workers to ignore.

Surveillance as a Flashpoint

The internal monitoring dispute centers on what Meta calls the Model Capability Initiative, a mandatory program on U.S. work laptops that captures employee activity to help train and evaluate A.I. tools. Workers who oppose the program have argued that it goes well beyond ordinary workplace software, turning routine labor into a stream of data for machine learning without meaningful consent.

An engineer’s internal protest against the practice has circulated widely inside the company, becoming a rallying point for broader frustration. Employees in the United States have organized around the issue, and workers in Britain have begun doing so as well, in part out of concern that the policy could spread beyond the U.S.

The scope of that organizing remains unclear. Organizers have not publicly disclosed how many workers have signed petitions or joined internal efforts. Nor is it known whether Meta intends to expand the tracking program into other jurisdictions, where worker-monitoring rules can be more restrictive.

Still, the backlash has exposed a basic trust problem. For many employees, the issue is not only that Meta is tracking how they work, but that the tracking appears tied to the company’s drive to build more powerful workplace A.I. systems at the same time it is reducing head count.

That dynamic has sharpened fears long voiced by labor advocates and workplace researchers: that A.I. may be introduced less as a tool to ease work than as a means to intensify it — gathering more data from employees, automating portions of their jobs and concentrating the gains in management and shareholders rather than in workers themselves.

The Cost of “Efficiency”

Meta has spent the last several years remaking itself around efficiency. After a period of aggressive expansion, the company moved into cost-cutting mode, reducing staff, flattening teams and demanding faster output from those who remained. The strategy has pleased investors and helped produce stronger margins. But employees say it has also left the company more anxious and less cohesive.

The coming layoffs threaten to deepen that mood. Cutting 8,000 jobs while holding thousands of open positions vacant suggests not simply a one-time reduction, but a reordering of what kinds of work Meta wants to prioritize as it channels more money into computing power, data centers and A.I. development.

Whether the May layoffs are an isolated measure or part of a broader restructuring tied to Meta’s A.I. ambitions is still uncertain. But for workers, the message is already resonant: the company is spending at historic levels on technology while asking employees to accept greater instability and oversight.

Why This Matters Beyond Meta

The dispute inside Meta arrives at a pivotal moment for white-collar work. As A.I. systems become more embedded in offices, coding workflows and knowledge work, employers are increasingly seeking data on how people perform their jobs. That can include not only the outputs workers create, but also the minute-by-minute traces of how they create them.

Meta’s case crystallizes the stakes. Here is one of the world’s most influential technology companies, flush with cash, cutting thousands of jobs and simultaneously collecting employee behavioral data to train better A.I. models. The question raised by that combination is larger than any one company: Who benefits when workplace activity becomes training material for automation?

For now, many of the answers remain unsettled. It is unclear whether worker opposition inside Meta will force changes to the monitoring program. It is unclear whether regulators, particularly in places with stricter privacy and labor protections, will take interest. And it is unclear whether the company’s work force, already strained by years of upheaval, can absorb another blow without further erosion of trust.

What is clear is that Meta’s internal conflict is no longer just a human-resources problem. It has become an early and unusually vivid test of how the A.I. economy will treat the people whose labor helps build it.

Sources

Further reading and reporting used to add context:

AI News

At Meta, Profit and Surveillance Deepen Worker Unrest