Meta Implements AI Training Through Employee Activity Monitoring

Employee Behavior Data for AI Development

Meta has begun collecting detailed records of how its staff engage with workstations, including every keystroke and mouse movement, to refine its AI systems. The initiative, introduced on Tuesday, operates through internal software and hardware, capturing user interactions to fuel machine learning processes. A company representative emphasized that the goal is to provide authentic examples of human-computer usage, stating, “Our models require real-world data to understand how people perform routine tasks.” They added that the tool includes measures to secure private information.

Reactions from Staff

Despite assurances, some employees express unease. One unnamed worker described the practice as “a stark reminder of a bleak future,” linking it to fears of further workforce reductions. “It feels like the company is pushing AI into every aspect of our lives,” they remarked. A former employee echoed these sentiments, calling the monitoring “another step in forcing AI acceptance across all departments.” This comes as Meta continues to streamline its operations, with recent layoffs and a hiring freeze.

Workforce Reductions and Strategic Shifts

Meta has already cut approximately 2,000 positions this year, though the latest moves suggest a larger scale of job losses. The company’s job portal, which once listed around 800 roles in March, now showcases only seven openings. The spokesperson from Meta declined to elaborate on the reduction of job postings or future personnel plans. Meanwhile, the firm aims to invest $140 billion in AI by 2026, nearly doubling its previous year’s allocation.

AI Expansion and Key Projects

According to Reuters, the tracking system is known as the Model Capability Initiative (MCI). The BBC was informed that while employee data was previously accessible, this tool marks a new focus on using it for AI improvement. In January, Zuckerberg highlighted 2026 as “the year AI transforms how we work,” noting that tasks once requiring large teams can now be handled by individuals. Last month, the restructured Meta Superintelligence Labs unveiled the AI model Muse Spark, which is expected to benefit from the new data streams.

“This company has become obsessed with AI,” said a current employee. “Having our smallest actions used for training feels very dystopian.”

“The tracking tool is just the latest way they’re shoving AI down everyone’s throat,” remarked a former employee.

Meta’s recent acquisition of nearly half of Scale AI for $14 billion underscores its commitment to expanding AI capabilities. By integrating data-labeling experts from the acquired firm, the company is accelerating its efforts to develop advanced models. As the workforce shrinks and AI investment rises, employees remain wary of the evolving landscape within the tech giant.