Meta Platforms Inc. began collecting comprehensive employee activity data this Tuesday, logging keystrokes and mouse clicks to train its artificial intelligence models, a move confirmed by company communications. This internal directive arrives as the tech firm navigates substantial job cuts and a near-total hiring freeze, intensifying concerns among its workforce, according to a former Meta employee who spoke to the BBC. The company states the data collection serves solely for AI development, not surveillance.
Meta Platforms Inc. deployed a new internal system this week, named Model Capability Initiative (MCI), designed to record granular employee interactions with company computers and applications. The tool, first reported by Reuters, captures data such as keystrokes and mouse movements. This information will feed directly into Meta's artificial intelligence training models.
The company informed its employees about the mandatory system on Tuesday, April 22, 2026. A Meta spokesman told the BBC that these models require "real examples of how people actually use them" to build effective AI agents for everyday tasks. He asserted the data is not used for other purposes and includes safeguards for sensitive content.
This operational shift arrives amid a period of deep unease within Meta's ranks. The company has already eliminated approximately 2,000 positions this year through various smaller rounds of layoffs. Many employees anticipate further, more substantial job reductions in the coming months, a sentiment previously reported by the BBC.
The visible signs of this contraction are stark. Last month, Meta implemented a partial hiring freeze. The company's public job board, which listed around 800 open positions in March, now displays only seven available roles.
The Meta spokesman declined to comment on the reduction in job listings or any future layoff plans. One Meta employee, who requested anonymity due to fears of professional reprisal, described the tracking initiative as "very dystopian." This individual told the BBC that the company has developed an "obsession with AI." Another person, who recently departed Meta, characterized the tracking tool as "just the latest way they're shoving AI down everyone's throat." These internal reactions reveal a disconnect between Meta's stated purpose for the data collection and the perception among its workforce, who view it through the lens of heightened job insecurity. Follow the leverage, not the rhetoric; the power dynamic is clear.
Mark Zuckerberg, Meta's co-founder and chief executive, has positioned AI as the company's central strategic priority. In January, Zuckerberg declared 2026 would be "the year that AI dramatically changes the way we work." He further noted at the time that "projects that used to take big teams now be accomplished by a single, very talented person." This vision, while ambitious, directly informs the current workforce anxieties. The math does not add up for a growing workforce when executive statements point to individual AI capabilities replacing large teams.
Meta plans to invest roughly $140 billion into AI projects throughout 2026. This figure represents nearly double the amount it dedicated to the technology in 2025. This substantial capital allocation underscores the company's aggressive pivot.
Part of this aggressive expansion included the acquisition of Scale AI in 2025 for more than $14 billion (£10.3 billion). Scale AI, a data-labeling firm, brought its executives into Meta to accelerate the development of the company's AI models and tools. The first major product from Meta's restructured Meta Superintelligence Labs group, the AI model Muse Spark, launched last month.
The data gathered from the new employee tracker is intended to refine and train subsequent AI models emerging from this lab. Historically, companies have sought to optimize employee productivity through various monitoring methods, from time clocks to email tracking. The current initiative represents an evolution of this trend, leveraging advanced AI capabilities to extract granular insights from daily work patterns.
This is not simply about tracking hours; it is about understanding the cognitive and operational sequences of human interaction with software. Here is what they are not telling you: The goal is to replicate and automate those sequences. Companies like Amazon have long utilized algorithmic management to monitor warehouse workers, but extending this level of detailed tracking to knowledge workers for AI training marks a significant escalation in corporate oversight.
The implications of such pervasive data collection extend beyond immediate job security. Privacy advocates, such as the Electronic Frontier Foundation, frequently raise concerns about the scope and retention of employee data, even when companies assert benign intentions. While Meta claims safeguards are in place to protect sensitive content, the sheer volume and detail of the information being logged could potentially create unforeseen vulnerabilities or misuse scenarios.
The line between training data and surveillance blurs when every keystroke is recorded. This move by Meta sets a precedent for how large technology firms might seek to extract value from their own human capital, transforming employee activity into a raw material for AI development. This aggressive push for AI dominance, fueled by internal employee data, carries significant strategic implications for Meta.
By rapidly developing proprietary AI, the company aims to reduce reliance on external tools and talent, potentially lowering long-term operational costs. However, it risks alienating its remaining workforce, fostering a culture of distrust that could hinder innovation and retention. The balance between technological advancement and human capital management proves precarious. - Meta is tracking employee keystrokes and mouse clicks for AI model training. - This mandatory internal data collection coincides with significant layoffs and a near-total hiring freeze at the company. - Meta plans to invest $140 billion in AI in 2026, nearly doubling its 2025 expenditure. - Employees express concerns about the "dystopian" nature of the tracking amid job insecurity.
The coming months will reveal the true impact of Meta's AI-driven strategy. Watch for further announcements regarding new AI product launches from the Meta Superintelligence Labs group. Observers will also closely monitor Meta's employment figures and any shifts in its hiring posture.
The tech industry will be watching to see if other major players adopt similar internal data collection strategies to accelerate their own AI development, setting a new standard for corporate data utilization and employee oversight.
Key Takeaways
— - Meta is tracking employee keystrokes and mouse clicks for AI model training.
— - This mandatory internal data collection coincides with significant layoffs and a near-total hiring freeze at the company.
— - Meta plans to invest $140 billion in AI in 2026, nearly doubling its 2025 expenditure.
— - Employees express concerns about the "dystopian" nature of the tracking amid job insecurity.
Source: BBC News
