[Exclusive] Meta to Back Yann LeCun’s New AI Venture with Strategic Compute Support
He plans to build a startup focused on Advanced Machine Intelligence (AMI).
Meta Platforms is preparing to invest in the new AI startup founded by Yann LeCun, the company’s outgoing chief AI scientist, in a move that could include providing Graphics Processing Units (GPU) access and other technical resources to support the research initiative, sources familiar with the matter told The Left Shift.
In October, The Left Shift reported that LeCun is unhappy with the recent restructuring at Meta and could exit the company very soon.
LeCun, widely regarded as one of the “godfathers” of deep learning and a Turing Award winner, officially announced his departure from Meta, where he pioneered foundational research and co-founded Facebook AI Research (FAIR).
He plans to build a startup focused on Advanced Machine Intelligence (AMI) — AI systems that can understand the physical world, maintain persistent memory, reason, and plan complex actions beyond current large language models. Meta has said it will partner with his new venture, reflecting ongoing collaboration even after his departure.
Last month, LeCun confirmed that Meta will be a partner in the new company, signaling a continued relationship despite his departure. Now, The Left Shift has learnt that there has been discussion within the Meta hierarchy to support Yan LeCun's startup with GPU support in return for equity in the startup.
Over the years, Meta has been aggressively building out its AI compute infrastructure to support its large-scale AI research and product development, particularly around its Llama family of language models and broader generative AI efforts. The company has deployed massive GPU clusters, each featuring tens of thousands of GPUs for training and running large neural networks.
Reports also indicate LeCun’s company may be based in Europe, particularly Paris, tapping into the region’s academic talent and supportive AI ecosystem. Placing the startup in Europe could also align with broader continental efforts to build sovereign AI capabilities and retain research leadership.
Comments ()