Meta has cut roughly 600 roles from its Superintelligence Labs, impacting infrastructure, FAIR-style research, and product teams. Mark Zuckerberg and Chief AI Officer Alexandr Wang said the move is meant to boost agility, focusing “load-bearing” work in fewer hands.
The elite TBD Lab remains untouched, and selective hiring continues. Still, the layoffs raise concerns about efficiency, the AI bubble, and the human cost of big tech’s pursuit of GenAI.
After months of aggressive tech investment, Meta AI appears to be consolidating its teams. Other tech giants are doing the same: Intel, Microsoft, Google, and Amazon have all reduced headcount in 2025, reflecting an industry-wide effort to streamline AI projects.
The decision to reduce the workforce by 600 people is aimed at making corporate processes faster while placing more accountability on fewer people. Smaller, more focused teams can be quicker, but analysts from the industry are cautioning that this will also increase pressure on the people left behind.
Analysts from the industry are observing that focusing resources on smaller teams can enhance decision-making and speed. This acts as a hindrance for the remaining staff, creating stress and lowering the variety of ideas in the room.
The Meta AI layoffs come as doubts grow over artificial intelligence investments. According to MIT's August 2025 "GenAI Divide" report, nearly 95% of corporate GenAI pilots yielded no significant return.
The difference between the hype and the reality reveals the perils of the AI bubble, suggesting that restructuring aimed at making companies more efficient could be more about the surface than the actual impact.
Meta continues to invest in AI, but the human cost is clear. Fewer people can mean faster execution, but it also limits collaboration and creativity. Swapping humans for hope isn’t transformation—it is cosmetic. The Meta AI layoffs illustrate a broader tension: the race for superintelligence versus the practical challenge of building it with real people.
Until corporate GenAI can produce consistent results, the scenario of big tech strongly pushing artificial intelligence growth is still ambitious and uncertain. This shows that the path to superintelligence is longer and more dependent on human resources than what corporate myths usually imply.