Meta employees in the US and UK are pushing back against company software that records their keystrokes and mouse activity, with an internal petition and a viral engineer’s post fueling growing unrest inside the company in May 2026.
Meta began installing the tool — which it calls the Model Capability Initiative — on US employees’ laptops last month. The software records screens during use of certain applications, capturing mouse movements, button clicks, and dropdown menu navigation. The stated goal is to collect real-world examples of how people use computers, data that can be used to train AI systems to navigate computers without human supervision.
An engineer’s internal post opposing the program was seen by nearly 20,000 coworkers. “Selfishly, I don’t want my screen scraped because it feels like an invasion of my privacy,” the engineer wrote. “But zooming out, I don’t want to live in a world where humans — employees or otherwise — are exploited for their training data.” A separate petition circulating since last Thursday calls for an end to the program, stating that companies should not be “permitted to exploit their employees by nonconsensually extracting their data for the purposes of AI training.”
The backlash has spread beyond digital channels. Workers at Meta offices in California and New York have been posting physical flyers in cafeterias and communal areas directing colleagues to the petition. The company has removed some of the posters, though those placed in bathrooms have reportedly remained up longer.
The controversy has become a leading factor in what 16 current and former employees recently described to WIRED as record-low morale at the company. It is also driving a unionization effort at Meta’s UK offices, where employees are not currently subject to the tracking but fear the program could expand. Eleanor Payne of United Tech and Allied Workers, which is helping organize UK employees, called the number seeking to unionize “significant” and described the situation as “a breakdown of trust.”
While US employers generally have broad legal authority to monitor workers’ devices, using that data to build AI training datasets appears to be a new application of that latitude — one that Meta has continued to pursue despite weeks of internal protest. The company has not yet disclosed whether the data collected has produced results.
Source: WIRED