Yann LeCun, one of the most prominent figures in artificial intelligence, is leaving Meta. After 12 years with the tech giant, the Chief AI Scientist announced he will launch his own startup. He plans to remain at Meta until the end of the year before shifting his full attention to his new venture.
LeCun joined the company in 2013 to establish its Fundamental AI Research (FAIR) lab. He says his new independent company will continue his work on “Advanced Machine Intelligence” (AMI). His goal is to move beyond today’s chatbots and create systems that genuinely understand the physical world, possess persistent memory, and can plan complex sequences of actions.
“Pursuing the goal of AMI in an independent entity is a way to maximize its broad impact,” LeCun wrote in a post on Threads. He added that his startup will partner with Meta, though it will operate as a separate entity.
The departure follows a period of evident friction regarding the future of AI development. LeCun is a vocal critic of the current hype surrounding Large Language Models (LLMs). He recently argued that simply making these models larger will not achieve human-level intelligence and even advised aspiring researchers to “absolutely not work on LLMs.”
Meta, however, has pivoted aggressively toward LLMs to compete with rivals like OpenAI and Google. The company recently restructured its AI division, appointing Alexandr Wang as Chief AI Officer and forming a new “Superintelligence” unit led by Shengjia Zhao. Reports suggest LeCun struggled to secure resources for his long-term research as the company focused on immediate commercial products.
Despite these strategic differences, LeCun expressed gratitude to Mark Zuckerberg and other leaders for backing the FAIR lab for over a decade. He now leaves the corporate hierarchy to chase the next breakthrough in machine reasoning.
