Meta’s AI overhaul: 600 jobs cut as Wang takes the helm

Meta is laying off about 600 employees from its artificial intelligence division as part of a broader effort to streamline operations and strengthen the leadership of its recently appointed Chief AI Officer, Alexandr Wang, CNBC reported.

The layoffs affect staff across Meta’s AI infrastructure, its Fundamental Artificial Intelligence Research (FAIR) group, and product-linked roles, though the newly formed Superintelligence Labs—home to top hires recruited this year—has been spared, according to people familiar with the matter cited by CNBC. The cuts are intended to reduce what insiders described as a “bloated” organisation and to eliminate overlaps in computing and research functions that had caused internal friction.

Wang, who joined Meta in June following the company’s $14.3 billion investment in Scale AI, is seen as pivotal to Chief Executive Mark Zuckerberg’s push to accelerate AI development. The restructuring consolidates Wang’s authority over the company’s sprawling AI operations, aligning Meta’s strategy more closely with that of industry rivals such as OpenAI and Google. Axios was first to report the layoffs.

Within Meta, tensions had reportedly emerged between long-standing FAIR researchers and newer recruits tasked with building cutting-edge AI products under Wang’s supervision. The latest round of layoffs reflects an attempt to resolve that tension while reshaping the business around high-impact, product-driven research. Following the cuts, Meta’s Superintelligence Labs workforce now stands at just under 3,000 employees, the sources told CNBC.

Employees affected by the layoffs were notified this week, with their final working day set for 21 November. Meta has offered 16 weeks of severance pay plus two additional weeks for each year of service, minus the notice period. “During this time, your internal access will be removed and you do not need to do any additional work for Meta,” read a message seen by CNBC.

The job reductions come as Meta doubles down on AI investment amid rising costs. The company has already spent heavily on data infrastructure, talent acquisition, and partnerships. In July, Zuckerberg announced Meta’s second-quarter forecast with total 2025 expenses expected between $114 billion and $118 billion—a figure expected to rise further in 2026 due to AI-related spending.

The CEO’s frustration reportedly grew after the lukewarm reception of Meta’s Llama 4 models earlier this year, which fell short of developer expectations. Since then, he has moved decisively to centralise control under Wang and former GitHub CEO Nat Friedman, who co-lead the Superintelligence Labs.

Meta’s aggressive restructuring comes as it seeks to sustain its lead in AI-powered products across its platforms, from Facebook and Instagram to WhatsApp. The company recently unveiled a $27 billion partnership with Blue Owl Capital to fund its new Hyperion data centre in Louisiana—described by Zuckerberg as “large enough to cover a significant part of Manhattan’s footprint.”

Meta is set to report third-quarter earnings next week, where analysts will be watching closely for signs of whether the streamlining and renewed AI focus are beginning to deliver returns.