Meta has introduced a new internal system to monitor US employees’ computer use, drawing complaints over privacy and lack of choice.
The tool, known as the Model Capability Initiative (MCI), records mouse movements, clicks, keystrokes, and takes regular screenshots on work laptops. It tracks usage across more than 200 apps and websites, including Gmail, Google Chat, coding software, and Meta’s AI assistant. Meta reportedly says the goal is to collect real examples of computer use to train AI agents for office tasks. The programme runs on company devices, with no option for employees to disable it.
The decision has sparked backlash inside Meta. Flyers criticising the company as an “Employee Data Extraction Factory” have shown up in offices, and workers have circulated a petition opposing the tool. Some employees described it as “uncomfortable” and “dystopian,” and voiced concern they are feeding data into AI that could automate their jobs. Others noted the tool consumed large amounts of home internet data. Meta’s CTO confirmed opting out is not allowed on work laptops.
Privacy questions have also emerged. Meta states MCI only monitors activity on work devices and will not be used for performance evaluations. However, if a US employee with MCI enabled chats or emails someone abroad, that interaction is captured. Experts warn this could clash with EU data-protection laws. Meta says the data is anonymised and privacy risks were addressed during development.
