Daijiworld Media Network - New Delhi
New Delhi, Sep 15: Meta has suspended an employee monitoring programme that collected computer activity to help train artificial intelligence models, following an internal data leak and growing opposition from employees, according to a report by Business Insider.
Meta’s Model Capability Initiative (MCI), launched within the company’s AI division earlier this year, collected data including employees’ keystrokes and mouse movements. The programme was designed to help AI models better understand how people use computers and perform routine tasks such as selecting options from dropdown menus and using keyboard shortcuts.
Meta announced the initiative to employees in April, saying AI models were strong in areas such as research and coding but still lacked some basic understanding of how humans interact with computers.
Employees were initially not given an option to opt out of the monitoring, triggering a backlash within the company. More than 1,800 employees later signed a petition demanding that the programme be scrapped.
Workers raised concerns that the monitoring could capture sensitive information, including passwords, immigration documents and credit card details. Some employees also feared that the data could eventually be used to develop AI systems capable of replacing human workers.
In June, Meta introduced an option allowing employees to temporarily pause the tracking for up to 30 minutes while handling sensitive tasks. However, employees opposing the programme said the measure was inadequate.
“We demand, and will continue to demand, that Mark and his leadership team scrap MCI entirely, immediately, forever,” employees said in their petition, referring to Meta CEO Mark Zuckerberg.
The programme was subsequently halted after a Meta engineer discovered that data collected through MCI could be accessed across the company, including private conversations.
Meta has said that MCI was suspended rather than permanently cancelled. The company has indicated that participation would be optional if the programme is brought back.
Meta has not disclosed how the data collected through the programme was used or whether it was ultimately used to train AI models.
According to Business Insider, the programme remained suspended two months after it was halted, leaving its future uncertain amid continuing concerns among employees over privacy, data security and the use of workplace monitoring for AI development.