This won’t make an already controversial AI training effort any more popular.
Meta has stopped using an AI training program that tracks its employees’ keystrokes and mouse movements. The company has suspended the Model Capability Initiative, not because of understandable resentment over employees being (almost) constantly monitored or potentially breaking privacy laws, but because it has led to leaks of internal data. business insider It was reported that sensitive data collected through MCI, including employees’ private conversations, performance data and transcriptions, was inadvertently made available to the entire Meta staff.
A spokesperson explained, “We have carefully designed this program with privacy safeguards in place, and although we have no indication at this time that any data was improperly accessed by Meta employees, we are putting it on hold while we investigate.” BI.
Despite this official line and previous statements that data collected by employees would be “strictly controlled”, it appears that Meta was not as on top of security as it claimed. This is the latest in a series of AI-related cybersecurity incidents for the company. The Meta representative issued a similar response in March when an agent AI took unexpected actions that also led Domino’s into a security breach. And earlier this month, the company was forced to respond after hackers exploited its AI customer service chatbot to hijack Instagram accounts.
<a href