Meta has suspended an employee-monitoring project after concerns emerged over how securely the information collected through the scheme was being handled.
The initiative, known internally as the Model Capability Initiative, was introduced about two months ago as part of Meta’s effort to improve its artificial intelligence systems. It recorded how employees used their workplace computers, including keystrokes, mouse movements, and clicks, with the aim of teaching AI tools to perform common office tasks more effectively.
The programme quickly became controversial inside the company. Employees questioned why such detailed monitoring had been introduced without broader consultation, while others raised concerns about who could see the data and whether private workplace information might be exposed.
Meta paused the scheme after discovering that some of the material may have been accessible to a wider group of employees than intended. The company is now reviewing its internal safeguards and investigating whether the information was viewed improperly.
A Meta spokesperson said the programme was “on pause for now” while the matter was examined.
“We have no indication at this time that any data was improperly accessed by Meta employees,” the spokesperson added.
Resistance had already been growing before the security issue came to light. Nearly 2,000 workers reportedly signed a petition calling for the initiative to be scrapped. Meta later allowed staff to stop the monitoring for periods of up to 30 minutes, although that concession did little to ease the backlash.
“That was just an attempt at damage control,” one employee said.
Another worker said many engineers accepted that Meta needed to strengthen its position against rivals such as OpenAI and Anthropic. However, the lack of consent remained a major source of anger, with the employee saying the tracking “was forced on us, there was no consent.”
The dispute has added to wider frustration within Meta as the company reorganises teams, cuts jobs, and directs more resources towards AI development. Its spending on data centres, computing power, and related infrastructure is expected to reach as much as $145 billion this year.
A former employee described Meta’s attempts to keep pace with competitors as “chasing your tail,” reflecting growing unease over the company’s direction and the pressure placed on its workforce.