Meta Cuts 600 AI Jobs as It Shifts Toward Superintelligence Labs

Meta Platforms is laying off approximately 600 employees from its artificial intelligence division in a surprising move that signals a dramatic shift in how the tech giant approaches AI development. The cuts target legacy research teams while the company doubles down on its ambitious superintelligence goals.

The layoffs, confirmed by Meta on October 22, 2025, affect the company’s Fundamental AI Research (FAIR) unit, AI product teams, and infrastructure divisions. Notably absent from the cuts? Meta’s newly formed TBD Lab, the elite team tasked with building the next generation of AI systems that could match or surpass human intelligence.

The Bureaucracy Problem That Sparked Mass Layoffs

According to an internal memo from Meta’s Chief AI Officer Alexandr Wang, the company’s AI efforts had become “overly bureaucratic.” Wang didn’t mince words about why these cuts were necessary.

“By reducing the size of our team, fewer conversations will be required to make a decision, and each person will be more load-bearing and have more scope and impact,” Wang wrote to employees.

The message is clear: Meta wants speed over size. The company believes smaller, more agile teams can move faster than large research groups bogged down by endless meetings and approval processes.

Affected employees received notifications on October 22 and face a termination date of November 21. During what Meta calls a “non-working notice period,” their internal access will be removed, though they can search for other positions within the company. Meta is offering 16 weeks of severance plus two weeks for every completed year of service, minus the notice period.

FAIR’s Fall From Grace

The cuts hit particularly hard at FAIR, Meta’s pioneering AI research lab. Founded in December 2013 by renowned AI scientist Yann LeCun, FAIR built a reputation as one of the world’s premier AI research institutions. The lab attracted top talent and published groundbreaking papers that advanced the field.

But times have changed. FAIR leader Joelle Pineau left Meta on May 30, 2025, after nearly eight years with the company. Her departure signaled trouble ahead for the research-focused unit.

The Decoder reports that CEO Mark Zuckerberg had grown frustrated with Meta’s AI progress. The company’s Llama 4 models, released on April 5, 2025, received only lukewarm responses from developers. While competitors like OpenAI, Anthropic, and Google continued releasing more powerful models, Meta seemed stuck in neutral.

Wang’s memo indicated that Meta will now try to integrate many of FAIR’s research ideas into larger projects run by the new TBD Lab. Translation? Pure research is out. Product-focused development is in.

The $14.3 Billion Bet That Changed Everything

To understand these layoffs, you need to rewind to summer 2025. That’s when Meta made a massive $14.3 billion investment in Scale AI, the data labeling startup co-founded by Alexandr Wang. The deal gave Meta a 49% stake in the company and brought Wang on board to lead Meta’s AI efforts.

Reuters notes that Meta went on an aggressive hiring spree after the Scale AI deal, recruiting top researchers from OpenAI, Google DeepMind, and Apple. The company spent hundreds of millions of dollars on talent acquisition, offering multimillion-dollar salaries to lure the best minds in AI.

But just months later, Meta paused hiring and announced this restructuring. What happened?

According to CNBC sources, the AI unit was considered bloated, with teams like FAIR and product-oriented groups competing for computing resources. When Meta created Superintelligence Labs and brought in new hires, the division inherited these oversized teams.

The result? Inefficiency, internal friction, and overlapping mandates that slowed progress.

Superintelligence Labs: Meta’s New AI Powerhouse

Meta established Superintelligence Labs on June 30, 2025, with one clear mission: build artificial general intelligence that can match or exceed human capabilities across all tasks. Zuckerberg calls this goal “personal superintelligence.”

The division has quickly become one of Meta’s most important and expensive bets. The New York Times reports that no cuts were made to TBD Lab, the team building superintelligence and managing Meta’s large language models.

This selective approach to layoffs sends a powerful message: Meta is betting everything on Wang’s team to deliver breakthrough AI systems. The company is moving away from open-ended research and toward shipping products that generate revenue.

Recent high-profile hires underscore this strategy. Meta brought in OpenAI research scientist Ananya Kumar and Andrew Tulloch, co-founder of Mira Murati’s Thinking Machines. These aren’t researchers who publish papers they’re builders who ship products.

The Financial Reality Behind the Cuts

Despite the layoffs, Meta remains incredibly profitable. The company reported $47.5 billion in revenue for the second quarter of 2025, up 22% from the previous year. So why cut 600 jobs?

The answer lies in Meta’s massive AI spending. The company expects total expenses for 2025 to reach $114 billion to $118 billion, with infrastructure costs being the largest driver. Building the data centers, acquiring AI chips, and powering the computing necessary for superintelligence requires enormous capital.

On Tuesday, Meta struck a $27 billion financing deal with Blue Owl Capital, the company’s largest-ever private capital agreement, to fund its biggest data center project. The deal allows Meta to shift much of the upfront cost and risk to external capital while retaining a smaller ownership share.

Saving several million dollars through layoffs might seem like a drop in the bucket compared to these massive investments. But it signals Meta’s commitment to efficiency and focus.

What This Means for AI Workers

Forbes reports that while these layoffs hit researchers hard, Meta is still hiring aggressively for its superintelligence efforts. The company wants builders, not just researchers.

This shift reflects broader changes in how tech companies think about AI work. Traditional research roles that publish papers are being replaced by engineering roles that ship products. The message to AI workers is clear: adapt or get left behind.

The irony isn’t lost on industry observers. AI researchers are losing their jobs because companies need to build better AI faster. The very technology these workers helped create is now reshaping their industry in ways they didn’t expect.

According to Lightcast research, AI-related job postings increased from just 55 in January 2021 to nearly 10,000 by May 2025. Yet Meta is cutting AI jobs. The disconnect highlights a fundamental shift: companies want specific AI skills focused on product development, not general research expertise.

Industry-Wide Implications

Meta’s layoffs are part of a larger wave hitting the tech industry. So far in 2025, there have been 571 layoffs at tech companies affecting 158,993 people 537 people per day. SiliconANGLE notes that Meta’s move bucks the trend of Silicon Valley companies aggressively hiring almost anyone with AI skills.

The layoffs are unprecedented in an AI industry that has been all about spending money. It raises an important question: will Meta’s competitors make similar moves in the months to come?

Some analysts suggest these cuts could trigger a broader reassessment of AI team structures across the industry. If Meta can achieve better results with smaller, more focused teams, other companies might follow suit.

Internal Tensions and Leadership Changes

Reports of tension in Meta’s AI divisions have surfaced before, including disagreements between Yann LeCun and the new management team. LeCun, who now reports to Wang, has reportedly clashed with the company over new publication rules and research direction.

These tensions reflect a fundamental disagreement about AI development philosophy. Traditional researchers like LeCun believe in open-ended exploration and publishing findings for the broader scientific community. The new leadership wants proprietary breakthroughs that give Meta competitive advantages.

Wang’s rapid rise to power from Scale AI CEO to Meta’s Chief AI Officer in just months has also created friction. Some longtime Meta employees resent outsiders coming in and dismantling teams they spent years building.

What Zuckerberg Really Wants

Earlier this summer, Zuckerberg hinted at these layoffs when he said Meta doesn’t need a “massive” team to make breakthroughs in AI. He emphasized that it’s better to go with “the smallest group of people who can fit the whole thing in their head, so there’s just an absolute premium for the best and most talented people.”

In July, Zuckerberg said the company would spend hundreds of billions of dollars to build several massive AI data centers for superintelligence. He’s betting that machines will soon match or surpass human capabilities and he wants Meta to get there first.

The CEO’s frustration with the Llama 4 reception triggered this major shake-up. He concluded that Meta’s existing AI structure couldn’t compete with rivals who were moving faster and achieving better results.

The Road Ahead

Meta expects most affected employees to find other positions internally because “we need their skills in other parts of the company.” The company has set up a “tiger team” of recruiters to help displaced workers find new roles through an expedited hiring process.

But make no mistake: this restructuring fundamentally changes Meta’s approach to AI. The company is moving from research-driven exploration to product-focused execution. From publishing papers to shipping features. From large teams to small, elite units.

Wang’s memo emphasized that “this by no means signals any decrease in investment. In fact, we will continue to hire industry-leading AI-native talent.” The key phrase? “AI-native talent.” Meta wants people who grew up building AI products, not researchers who transitioned from academia.

Meta is scheduled to report its third-quarter earnings results next week, when it may reveal more about the reasons for these layoffs and its AI strategy going forward.

The Bottom Line

Meta’s decision to cut 600 AI jobs while pursuing superintelligence sends a clear message: the company believes smaller, more focused teams can achieve breakthrough results faster than large research organizations. Whether this bet pays off remains to be seen.

For AI workers, the lesson is sobering. Even at the forefront of the most transformative technology in decades, job security isn’t guaranteed. The skills that got you hired yesterday might not be what companies need tomorrow.

The AI revolution is eating its own. Workers who spent years teaching machines to think are now being told they’re not moving fast enough. In the race to build superintelligence, even the AI experts aren’t safe.