Meta's AI Reorg Sparks Internal Revolt: Employee Outburst Exposes Morale Crisis Amid Rapid Restructuring

A routine internal Meta meeting earlier this week turned chaotic when an employee interrupted a livestreamed presentation with a profanity-laced tirade, calling themselves 'the company's bitch' and demanding leaders tell a senior AI executive he is 'a piece of shit.' The outburst, captured in a recording reviewed by WIRED, highlighted deep frustrations within Meta's newly formed Applied AI unit, established in March 2026 to support Superintelligence Labs research. The roughly 6,500 engineers and product managers in the unit have been reassigned to repetitive tasks like generating puzzles and coding challenges for AI model evaluation, leading multiple anonymous employees to describe the work as 'soul-crushing' and 'literally the gulag.'

This incident fits into a broader pattern of upheaval. In May 2026, Meta laid off approximately 8,000 employees while transferring another 7,000 into AI-focused roles as part of an aggressive pivot. Reports from Reuters and Bloomberg detail the scale, with Chief People Officer Janelle Gale announcing flatter structures and new AI organizations like Applied AI Engineering. Employees across divisions, including data centers and Instagram, have reported heightened stress.

Compounding tensions, over 1,600 employees signed a petition against a program monitoring U.S. workers' clicks, keystrokes, and screens to generate AI training data; WIRED notes the program was later scaled back. In an Instagram all-hands, Chief Product Officer Chris Cox acknowledged the 'brutal' environment, comparing it to 'running a marathon in the middle of a hailstorm' and stating AI 'is neither god, nor is it the devil.'

CEO Mark Zuckerberg addressed the chaos in a June 2026 internal memo, per Reuters, admitting 'we've made mistakes' in the AI workforce shift while promising stability and no further company-wide layoffs this year. TechCrunch and other outlets corroborate the Applied AI unit's rapid formation under Maher Saba and the resulting disillusionment.

These events reveal raw industry-wide fractures: the human cost of AI acceleration, ethical concerns over workplace surveillance for model training, and the demotivation of skilled engineers reduced to data-labeling roles. Mainstream coverage often glosses over such real-time sentiment, but the documented employee actions and executive admissions underscore systemic strains in Big Tech's AI race.