Meta's Former AI Chief LeCun Calls xAI a 'Failure,' Warns of AI Industry 'Bubble Explosion'

Yann LeCun, former Meta chief AI scientist and founder of AMI Labs, told CNBC that Elon Musk's xAI is "kind of a failure" that will be unable to compete at the frontier of AI with companies like OpenAI and Anthropic.

Key Criticisms of xAI

LeCun cited the departure of several co-founders from xAI over the past year as evidence of the company's struggles. "Elon is now in a position that is very, very difficult for him to kind of hire top people in AI, because he's kind of, you know, not behaved in sort of very good ways toward the previous team," LeCun said.

According to LeCun, xAI maintains "huge infrastructure" at its Colossus 1 and Colossus 2 data centers in Memphis, Tennessee, which it rents to other companies including Google and Anthropic "because that's the only way he [Musk] can recoup the cost."

In February 2026, Musk merged SpaceX with xAI in a deal that valued the combined entity at $1.25 trillion. However, SpaceX's AI segment, which includes xAI, posted a $2.5 billion operational loss in the three months ended March 31.

Industry-Wide Economics Warning

LeCun issued a broader warning about AI industry economics. "The prices are going up of those AI services, but the cost of running them is going down, but not nearly fast enough. And so all of those companies are losing money, and basically, the use for most people is funded by the investors. That can't go on for a very long right?" he said.

LeCun warned that labs like OpenAI and Anthropic "are going to have to increase prices, they're going to have to cut costs, or there's going to be a big bubble explosion."

OpenAI CEO Sam Altman reportedly acknowledged AI costs as a "huge issue" in a company livestream this month, noting that companies are now actively discussing their AI spending.

World Models vs. LLMs

LeCun has been a vocal critic of large language models (LLMs), the foundation of current AI systems. Instead, he advocates for "world models" that build understanding of how the real or simulated world works through objects, cause and effect, and actions.

"I personally don't think we're going to have generalized reliable agentic systems until they're based on world models," LeCun said. His company AMI Labs raised $1 billion in March 2026 at a pre-money valuation of $3.5 billion to pursue world models research.

LeCun acknowledged LLMs are useful for coding and math but noted "the cost of running those systems with this kind of performance is very high compared to the amount of money that users are ready to pay."

Ongoing Tensions

The comments renew a long-running dispute between LeCun and Musk. LeCun has previously criticized what he described as Musk's "conspiracy theories" on social media, while Musk has accused LeCun of being "out of touch with AI for a long time."

What This Means

LeCun's comments highlight two critical pressure points in the AI industry: the sustainability of current business models where investor funding subsidizes user costs, and fundamental questions about whether LLM-based architectures can deliver on promises of autonomous AI agents. His prediction of a potential "bubble explosion" echoes growing concerns about AI valuations as operational losses mount across leading labs. The economics criticism carries particular weight given LeCun's stature as a Turing Award winner and pioneer in deep learning, though his advocacy for world models also positions him as a competitor to the current LLM paradigm.