The internal environment in Meta is not at its best. Workers who do not have been fired were relocated to a new AI department described as “the gulag”. Additionally, the company recently installed software on workers’ computers that recorded everything they did. Well, they had to pause it because there was a leak that left the data exposed.
what has happened. They tell it in Wired. Meta has had to pause its employee tracking program due to an internal leak. The program, called Model Capability Initiative or MCI, was launched in April and was installed on all company computers. This software recorded screen content, keystrokes and clicks. It was not to control employees (or not only for that), but to train AI agents who, presumably, will one day replace them. As expected, employees were not amused, nor were they amused that a security issue left their data exposed to other workers.
The reaction. The notice came from an employee who detected that he could access the MCI data of other employees. According to Meta, the bug was solved in four hours, although the first solution was not permanent and they had to reinforce the data access block. During those hours, employees, who had already expressed their discontent with this program, filled the internal forums with criticism and Meta finally suspended the program. But be careful because it is not goodbye, but see you later. Speaking to Wired, AI research vice president Stephan Kasriel said that “we will reactivate MCI when we are confident in the effectiveness of our data protection controls.”
And now they go… The case adds to Meta’s long history with leaks and privacy issues. The clearest and most resounding was the Cambridge Analytica scandal in 2018, but there have been many morelike the massive data theft which affected a whopping 533 million users in 2021 or the use of facial recognition without permission in 2024. We recently learned that Meta’s glasses record even when we take them offmeaning contractors in Kenya have seen users having sex, going to the bathroom or accessing their bank accounts.
This questionable way of treating its users’ data and repeated scandals have earned the company a reputation for being unreliable when it comes to privacy. We now know that even employees themselves are vulnerable to this.
Fed up employees. As we said, to the employees They were not happy that the company installed this software on their computers. In fact, at least 1,600 employees had signed a petition in which, in addition to describing the program as invasive, they warned that it posed a risk from a security point of view. All this was happening at a time when Meta had announced layoffs, but had not yet communicated who those affected were. For a month, the staff worked without knowing if they were going to be fired, but the anguish did not end with the “execution.”
Meta laid off more than 8,000 employees and relocated another 7,000, who now work in the Applied AI department. Many high-level engineers have ended up here and describe the work as mechanical and uncreative. Now, in addition, they leak their private data from the program they were using to control them. It is not surprising that the internal environment is going through its lowest hours.
Image | Xataka with ChatGPT
In Xataka | Soon you won’t need to remember anyone’s name. Meta’s glasses will do it for you
GIPHY App Key not set. Please check settings