Meta’s 6,500-person Applied AI unit is facing internal revolt. Engineers forcefully reassigned to the division liken the experience to a “soul-crushing gulag”, citing forced ultimatums, monotonous training tasks, intrusive workplace surveillance, and chaotic management structures.
The Core Issues.
- Forced Conscription: Engineers were given a harsh ultimatum: join the AI unit or quit the company. Many internally refer to themselves as “draftees”.
- Menial Labor: Rather than high-level software development, these engineers are primarily tasked with generating coding puzzles and evaluation tests to train AI models—drudgery they find unfulfilling.
- Intrusive Monitoring: More than 1,600 Meta employees have signed a petition protesting an internal program that monitors their keystrokes and mouse clicks for AI training data.
- Chaotic Structure: In its initial months, the unit experienced severe structural issues, with up to 50 employees reporting to a single manager.
The Fallout.
- Open Revolt: Employee frustration has boiled over. During a livestreamed, company-wide presentation, an employee hijacked the audio to swear at a senior AI executive and demand others voice their complaints.
- Leadership Response: The environment is so tense that Chief Product Officer Chris Cox reportedly called it “brutal” on an internal call.
- Zuckerberg’s Memo: CEO Mark Zuckerberg addressed the unrest in an internal memo, acknowledging that recent organizational changes had “caused distress” and admitting the company had made mistakes.
The structural tension at Meta mirrors broader issues within Big Tech regarding the need for high-quality human-generated data to improve AI capabilities. Discussions surrounding the ethical and psychological toll of these reassignments can be found on platforms like Reddit and Digg.
