Meta’s long-time Chief AI Scientist Yann LeCun, one of the most influential figures in artificial intelligence, is reportedly leaving the company to start a new venture dedicated to building next-generation AI systems. The move marks a pivotal moment for Meta’s AI division, as LeCun has been one of the leading voices shaping the company’s research and strategy in artificial general intelligence (AGI).
LeCun, who joined Meta (then Facebook) in 2013, played a central role in developing the company’s AI research division, known as FAIR (Facebook AI Research). Under his leadership, FAIR became one of the most respected AI labs globally, contributing to advances in computer vision, deep learning, and self-supervised learning — technologies that now underpin much of the world’s AI innovation.
According to reports, LeCun’s upcoming startup will focus on creating AI systems that go beyond large language models, aiming to replicate the reasoning and learning capabilities of the human brain. His new company will reportedly pursue research into architectures that enable machines to understand the world more efficiently, bridging the gap between today’s generative models and future forms of artificial general intelligence.
LeCun, a Turing Award winner widely regarded as one of the founding fathers of deep learning, has often expressed skepticism about the current trajectory of AI. He has been vocal about the limitations of large language models like ChatGPT, which, in his view, excel at pattern recognition but lack fundamental understanding. His departure from Meta signals a shift toward pursuing what he has long described as “autonomous intelligence” — AI systems capable of reasoning, predicting, and planning in a manner closer to human cognition.
Focus on new-generation AI architectures
People familiar with the development said that LeCun’s startup will likely concentrate on developing energy-efficient neural architectures inspired by biological systems. While the company’s name and funding details have not yet been announced, early indications suggest it will attract strong backing from investors interested in long-term AI research rather than short-term commercial applications.
Meta’s spokesperson confirmed that LeCun will continue to serve as an advisor during the transition phase but will focus primarily on building his independent AI venture. “Yann has been instrumental in shaping Meta’s AI journey, and we are deeply grateful for his contributions,” the company stated in a brief comment.
LeCun’s exit comes at a time when major technology companies are racing to build advanced AI models. Meta has been doubling down on open-source AI research, positioning its Llama family of models as a counterweight to OpenAI’s proprietary systems. Meanwhile, LeCun’s approach to AI development emphasizes self-supervised learning and energy-based models, which he believes are essential for building more general-purpose intelligence.
A legacy of innovation and debate
Over his decade-long tenure at Meta, LeCun was instrumental in fostering academic collaborations and open research practices in AI. He also maintained his professorship at New York University, where he mentored several of today’s leading AI researchers. His influence extends far beyond Meta, shaping how the global AI community views machine perception and representation learning.
LeCun’s departure follows a broader reshuffling within the AI leadership at major tech companies. It mirrors recent moves by executives from Google DeepMind, Anthropic, and OpenAI, who have left to establish independent ventures focused on specialized AI applications. This trend reflects growing diversity in AI research directions, as scientists seek to balance commercial priorities with long-term innovation goals.
In previous statements, LeCun has emphasized that the next phase of AI will not be achieved by scaling existing transformer-based models but by creating systems that learn continuously and autonomously. His startup is expected to focus on this principle, designing AI that can form internal world models, learn causality, and make predictions about real-world events without constant retraining.
AI industry implications
Analysts believe LeCun’s decision could influence the broader research ecosystem, especially among academic and industrial AI labs pursuing AGI. “LeCun’s move highlights a growing desire among top researchers to explore alternative paths to general intelligence,” said one industry observer. “It may also spark renewed competition between independent research startups and large corporate AI labs.”
Meta, meanwhile, continues to expand its AI initiatives under Mark Zuckerberg and Chief Product Officer Chris Cox, focusing on integrating generative AI tools across Instagram, WhatsApp, and Facebook. The company recently introduced new AI-powered content creation and personalization features for users and advertisers.
As one of the field’s most prominent researchers, LeCun’s next chapter will be closely watched. His decision to step away from one of the world’s largest AI programs to build a new enterprise underscores both the promise and complexity of the global AI race — a race increasingly defined not just by scale, but by vision and principle.