Meta is slashing 600 jobs from its artificial intelligence unit.
Chief AI Officer Alexandr Wang announced the layoffs in an internal memo Wednesday. Workers across several groups will be affected: • AI infrastructure teams • Fundamental AI Research unit • Various product-related positions
Wang was hired in June as part of Meta's massive $14.3 billion investment in Scale AI. Now just months later, he's delivering layoffs.
Meta says the cuts are about reducing layers and operating more efficiently.
Here's what makes this weird: Meta is simultaneously firing AI workers while spending insane amounts on AI.
Just Tuesday, the company announced a $27 billion deal with Blue Owl Capital to build the Hyperion data center in rural Louisiana. Zuckerberg said it will be large enough to cover "a significant part of the footprint of Manhattan."
Meta also expects total expenses for 2025 to hit between $114 billion and $118 billion. The company already raised the low end of that range. And it warned that AI spending will push 2026 expenses even higher than 2025.
So why fire 600 AI employees while spending tens of billions on AI infrastructure?
CEO Mark Zuckerberg has been unhappy with Meta's AI progress.
The company's Llama 4 AI models launched in April to a lukewarm response from developers. That apparently didn't sit well with Zuckerberg, who's racing to keep up with OpenAI and Google.
After the Scale AI investment, Zuckerberg created a new unit called Meta Superintelligence Labs. It's staffed with top AI researchers and engineers, led by Wang and former GitHub CEO Nat Friedman.
The message seems clear: Meta wants elite talent working on breakthrough AI, not just bodies filling roles.
These layoffs are about restructuring, not scaling back.
Meta isn't giving up on AI. It's doubling down. But the company apparently thinks it has too many layers and not enough productivity.
This is similar to what other tech companies have done - fire middle management and focus resources on key projects with top talent.
The timing is notable. Meta reports third-quarter earnings next week. Announcing layoffs right before earnings suggests management wants to show Wall Street it's controlling costs while still investing heavily in AI.
Meta is playing a delicate game.
On one hand, it needs to convince investors that massive AI spending will pay off. Hence the $27 billion data center deal and promises of breakthrough research.
On the other hand, it needs to show operational discipline. Hence cutting 600 jobs from the AI division.
For the employees being laid off, this stings. You're working in AI - supposedly the hottest area in tech - at a company spending billions on AI. And you still get fired.
For investors, the question is whether Meta's AI strategy is working. Spending more money doesn't automatically equal better results. Sometimes you need fewer, better people rather than more bodies.
Zuckerberg's frustration with Llama 4's reception suggests Meta knows it's behind competitors in the AI race. Restructuring might help. Or it might just be shuffling deck chairs while OpenAI and Google pull further ahead.
Meta's earnings next week will be crucial. Investors will want to see progress on AI that justifies the massive spending. If Meta can't show meaningful AI wins soon, these layoffs might be just the beginning.