AI godfather Yann LeCun calls Meta billionaire AI chief Alexandr Wang ‘inexperienced’

"He learns fast, he knows what he doesn’t know," LeCun, 65, who served as Meta’s chief AI scientist before announcing in November 2025 that he was leaving to form his own startup, said of Wang, 29, in a recent interview with the Financial Times.

"There’s no experience with research or how you practice research, how you do it. Or what would be attractive or repulsive to a researcher."

Yann LeCun, often described as one of the "godfathers of AI." Photo courtesy of LeCunn's X

In June 2025, Meta invested $15 billion in data-labeling startup Scale AI and recruited its chief executive and co-founder, Wang. Once briefly the world’s youngest self-made billionaire, Wang now has an estimated net worth of $3.2 billion, according to Forbes.

He was placed in charge of Meta’s artificial intelligence push and its research unit, known as TBD Lab, which is tasked with developing next-generation frontier AI models. The move briefly made Wang become LeCun’s direct superior. However, LeCun said Wang did not actively direct his work. "You don’t tell a researcher what to do," LeCunn said. "You certainly don’t tell a researcher like me what to do."

In the same interview, LeCunn said Meta chief executive Mark Zuckerberg grew frustrated after disappointing progress on Llama, the company’s flagship open-source AI model, and "basically lost confidence in everyone who was involved" after Meta was accused of gaming benchmarks to make its Llama 4 model appear more capable than it was, according to Business Insider.

Alexandr Wang, 29, Meta's chief AI officer. Photo from X

Zuckerberg "basically sidelined the entire Gen AI organization," Lecun said, describing a sharp internal reset following the controversy.

LeCun has repeatedly argued that large language models are fundamentally limited and that unlocking more advanced forms of artificial intelligence requires a different approach. That belief underpins his new startup, reportedly named Advanced Machine Intelligence, which reflects the research direction he has long promoted as a more viable alternative to LLMs.

Nabla, a health tech AI startup that partnered with LeCun’s company last December, said in a press release that LLMs, unlike world models, "still face some structural constraints, including hallucinations, non-deterministic reasoning and limited handling of continuous multimodal data, which make autonomous decision-making challenging," CNBC reported.