Meta’s Chief AI Scientist to Exit and Launch Startup

Meta’s top AI scientist, Yann LeCun, is reportedly preparing to leave the company to launch his own startup.

This could mark the most significant exit yet from Mark Zuckerberg’s ambitious campaign to build “superintelligence” at the social media giant.

LeCun, 65, a French-American computer scientist and one of the world’s most influential AI researchers, has informed colleagues that he plans to depart in the coming months, according to a Financial Times report. The Turing Award laureate, often described as a “godfather of deep learning,” is said to be in early talks to raise funding for his new venture.

His exit underscores deepening tensions within Meta over the future of AI development, as Zuckerberg pivots aggressively toward rapid commercialization of AI tools while de-emphasizing the open-ended, exploratory research that LeCun championed for more than a decade.

The end of an era

LeCun has led Meta’s Fundamental AI Research Lab (FAIR) since its founding in 2013. FAIR became one of the world’s most respected academic-style research units, producing groundbreaking work in neural networks, computer vision, and machine learning theory.

However, Meta’s strategic direction has shifted sharply in recent years. Facing growing competition from OpenAI, Google, and Anthropic, Zuckerberg has instructed his teams to prioritize the rollout of market-ready AI products over the slower pursuit of scientific breakthroughs. The pivot reflects a belief that Meta must rapidly establish itself as a leader in generative AI or risk being left behind in the sector’s explosive growth.

LeCun’s research at FAIR has focused on what he calls “world models” — AI systems capable of understanding the physical world by learning from spatial and video data rather than relying solely on text. He has repeatedly argued that these systems, though still years away, are essential for developing machines that can truly reason like humans. LeCun has warned that large language models, or LLMs, which underpin popular chatbots such as ChatGPT, are “useful” but fundamentally limited in their ability to plan and think abstractly.

Zuckerberg’s drive toward “superintelligence”

Zuckerberg’s new direction is embodied in the creation of Meta’s “superintelligence” division — a high-stakes initiative designed to accelerate the company’s ability to produce AI systems that can rival or surpass human intelligence in specific tasks.

Over the summer, Meta spent more than $14 billion to acquire a 49% stake in Scale AI and hire its founder, Alexandr Wang, to lead the superintelligence team. The move was both a bold bet and a signal of Zuckerberg’s willingness to spend heavily to catch up with his competitors. LeCun, who once reported directly to Meta’s chief product officer Chris Cox, now reports to Wang, a 28-year-old entrepreneur known for his data-labeling empire.

Zuckerberg has also poached engineers and scientists from Google, OpenAI, and Anthropic with compensation packages exceeding $100 million. In July, Meta hired Shengjia Zhao, one of the co-creators of OpenAI’s ChatGPT, as chief scientist of the new division. The recruitment spree has fueled internal friction, with some veteran Meta researchers frustrated by the ballooning pay disparities and a culture increasingly focused on speed over scientific rigor.

Investor pressure and market turbulence

LeCun’s planned exit comes amid mounting investor unease over Meta’s AI spending. Shares fell 1.2% following reports of his departure and have dropped more than 12% since late October, when Zuckerberg warned that AI investments could surpass $100 billion next year. That announcement wiped roughly $240 billion from Meta’s market value in a single week.

Analysts say the sell-off reflects growing skepticism that Big Tech’s multibillion-dollar push into AI will yield near-term profits. Wall Street has begun to question whether the hype around generative AI — which has driven soaring valuations across the sector — can translate into sustainable business models. For Meta, whose advertising-driven revenues remain under pressure, the stakes are especially high.

Zuckerberg has promised that Meta’s AI systems will soon automate much of the company’s coding and product design work. But for many investors and researchers alike, such projections appear optimistic given the company’s uneven track record — including the underwhelming performance of its Llama 4 model and the lukewarm reception to its Meta AI chatbot.

If LeCun’s startup successfully advances his long-term vision, it could represent a new paradigm in AI — one that prioritizes perception and common-sense reasoning over statistical prediction. Such an approach might ultimately bridge the gap between today’s chatbots and the kind of general intelligence that AI pioneers have long imagined.

Drama in the AI division

LeCun’s exit caps a turbulent year of turnover and restructuring at Meta. In May, Joelle Pineau, vice president of AI research, left to join the Canadian startup Cohere. Last month, the company laid off about 600 employees from its AI division in an effort to cut costs and speed up product delivery.

Despite these challenges, Zuckerberg remains steadfast in his conviction that AI will redefine Meta’s future. “Superintelligence,” he has said, will underpin everything from digital assistants to advanced creator tools. But the loss of LeCun — one of the field’s foundational thinkers — raises questions about whether Meta’s pursuit of rapid deployment will come at the expense of innovation itself.

Elon Musk, a man who likes to offer opinions almost on a 24/7 basis, warned of a “supersonic tsunami” of AI that could soon wipe out desk jobs.