Meta will track employee keystrokes and clicks to train AI models
Meta will begin logging employees' keystrokes, mouse clicks, and other computer activity to train artificial intelligence systems. The company told workers on Tuesday that a new tool called the Model Capability Initiative will run on Meta computers and internal apps, capturing how people work to feed into AI development.
A Meta spokesman said the tracking serves a specific purpose: building AI agents that can help people complete everyday tasks. "If we're building agents to help people complete everyday tasks using computers, our models need real examples of how people actually use them," he told the BBC.
The company said safeguards protect sensitive content and the data is not used for other purposes. But employees have raised concerns about the move, particularly given the timing.
Employee concerns amid job cuts
Meta has already cut roughly 2,000 employees this year in smaller rounds. Workers expect deeper layoffs in coming months, and the tracking announcement has amplified unease.
One Meta employee said the initiative feels "very dystopian" when combined with expected job losses. "This company has become obsessed with AI," they told the BBC on condition of anonymity.
Another former employee described the tool as "just the latest way they're shoving AI down everyone's throat."
Job listings on Meta's careers site dropped sharply. The site hosted about 800 positions in March; it now shows just seven. Meta declined to comment on the reduction or planned cuts.
Part of broader AI spending push
The tracking tool fits Meta's aggressive AI strategy. Chief Executive Mark Zuckerberg has pledged to increase AI spending significantly, positioning the company at the forefront of the technology.
Meta plans to spend roughly $140 billion on AI in 2026, nearly double its 2025 investment. Last year, the company took a controlling stake in Scale AI, a data-labeling firm, for $14 billion and brought some of its executives into Meta to help build AI models.
Zuckerberg said in January that 2026 will be "the year that AI dramatically changes the way we work." He noted that projects once requiring large teams can now be completed by single talented individuals.
Data collected from the employee tracking tool will help train new AI models being developed by Meta's Superintelligence Labs group, which launched the Muse Spark AI model last month.
What this means for HR leaders
Meta's approach raises questions for HR professionals about employee monitoring, data privacy, and workforce management in an AI-driven environment. HR leaders should understand both the technical capabilities and the employee relations implications of such tracking systems.
For those managing AI implementation in their organizations, AI for Human Resources offers practical guidance on workforce monitoring and employee data management. HR executives considering broader AI strategy should review the AI Learning Path for CHROs, which covers workforce analytics and talent management alongside AI implementation decisions.
Your membership also unlocks:
