Meta's AI Unit: A 'Gulag' of Discontent? Engineers Spill Shocking Truths

Meta is reportedly experiencing significant internal unrest and low employee morale, exacerbated by a series of ongoing layoffs and the company's substantial investment in artificial intelligence. This challenging environment has particularly affected the company's Applied AI team, which is rumored to be on the verge of revolt.

The internal tensions came to a head during a recent employee-only livestreamed presentation, which was dramatically interrupted by an expletive-laden outburst. An individual demanded that attendees convey a message to a senior Meta AI executive, accusing them of being “a piece of sh*t,” an incident that reportedly left one presenter covering their face in dismay. This public display is said to reflect the deep-seated anger and frustration within the Applied AI unit, a three-month-old division comprising approximately 6,500 engineers and product managers tasked with advancing Meta's AI research ambitions.

Many employees within this unit reportedly learned of their reassignment through a surprise email, a process described by one “draftee” on Reddit as “quite random.” An internal announcement, reviewed by Business Insider, explained the rationale behind these forced transfers: Meta's AI models were still incapable of outperforming humans in technical tasks like coding. The company stated a need to “train our models on real examples” to help AI agents understand how people complete everyday tasks using computers.

CEO Mark Zuckerberg elaborated on the decision to utilize internal employees rather than external contractors in a leaked audio recording from an internal meeting. He cited Alexandr Wang, who sold his data-labeling startup Scale AI to Meta for $14.3 billion and now heads Meta Superintelligence Labs, as someone with extensive knowledge of the data-labeling sector. Zuckerberg candidly added that the average Meta employee possesses “significantly higher” intelligence than third-party contractors, making them a more suitable choice for the task. However, employees describe being given no real choice but to “join or quit,” leading many to self-identify as “draftees.”

The nature of the work assigned to these draftees has been a major source of discontent. Their primary task involves generating puzzles and coding problems specifically to train AI models. Employees have voiced strong negative sentiments about this work, with one telling Wired it was “literally the gulag” and another describing it as “soul-crushing.”

The morale issues are not confined solely to the Applied AI group. A broader sense of dissatisfaction permeates the company, evidenced by over 1,600 Meta employees reportedly signing a petition protesting a program that monitors their clicks and keystrokes for AI training data. The prevailing mood across Meta is reportedly so dark that Chief Product Officer Chris Cox felt compelled to address the “brutal” environment during a call with employees this week.

Earlier reports indicate that the Applied AI team is led by Maher Saba, a 12-year Meta veteran who previously served as a vice president in Reality Labs—the division that incurred $83 billion in losses on the metaverse before Meta shifted its focus to AI. The new organization reports directly to Meta CTO Andrew Bosworth. Initially, the unit faced structural challenges, with up to 50 employees reporting to a single manager, which likely contributed to operational inefficiencies and employee frustration.

In response to the widespread internal distress, CEO Mark Zuckerberg reportedly issued an internal memo on Friday. In it, he acknowledged that recent changes had “caused distress” and admitted that the company had made mistakes, which it plans to address. According to Wired, Zuckerberg concluded his memo by reaffirming that “Meta’s north star is to be the best place for the most talented people in the world to make an impact.”