Yann LeCun Calls Elon Musk’s xAI a ‘Failure’ Yann LeCun has escalated his long‑running skepticism about today’s AI race by declaring Elon Musk’s xAI “kind of a failure,” turning a technical rivalry into a pointed critique of one of the sector’s highest‑profile startups.

Early doubts and staffing turmoil

In an interview this week, LeCun argued that xAI will not be able to compete at the frontier with giants like OpenAI and Anthropic, basing his assessment largely on people rather than technology. He said the company’s “founding team has left, or was fired,” adding that Musk now finds it “very, very difficult” to hire top AI talent because he has “not behaved in sort of very good ways” toward the previous team.

Reports from earlier in the day had already framed LeCun’s comments as branding xAI a “failure,” linking the alleged hiring struggles directly to those early departures.

Economic pressure and infrastructure strategy

LeCun then broadened his criticism from management to economics. He noted that xAI has built “huge infrastructure” that it rents to others “because that’s the only way he can recoup the cost.” According to coverage of the deal structure, xAI’s Colossus data centers in Memphis now lease capacity to rivals Anthropic and Google, with Google alone reportedly paying SpaceX about $920 million a month, as part of a structure in which Musk merged SpaceX with xAI, valuing the combined entity at $1.25 trillion while its AI segment posted a $2.5 billion quarterly operating loss.

From xAI to an AI bubble warning

LeCun used xAI’s struggles as a springboard to warn that much of the industry faces similar financial strain. AI services are becoming more expensive to run than labs can profitably charge, he argued, with investors effectively subsidizing users: “That can’t go on for very long … there’s going to have to increase prices, they’re going to have to cut costs, or there’s going to be a big bubble explosion.”

His broader skepticism is consistent with his public stance that current large language models are a dead end and that new “world models” will be needed for AI systems more grounded in reality. On social media, LeCun amplifies arguments that today’s generative systems, like ChatGPT, Claude, or Gemini, remain “extremely skilled” at language yet “not connected directly to physical reality,” highlighting a gap he believes future AI must close.

Continue reading https://foxvector.com/stories/019edf9b-42c9-08dc-7309-1a4dcadb2749