In late 2013, Facebook established Facebook AI Research, known as FAIR, with Yann LeCun as its founding director. Meta’s own ten-year retrospective recounts that the lab was built when the company decided to “assemble a team of some of the most talented researchers in the nascent field of deep learning,” with Mark Zuckerberg personally recruiting researchers at the NeurIPS conference. LeCun, a pioneer of convolutional neural networks, joined as VP and Chief AI Scientist while keeping a position at New York University.
FAIR was unusual among corporate labs for its commitment to open research. Meta describes a philosophy of “advancing the state of the art in AI through open research,” sharing “papers, code, models, demos or responsible use guides” rather than keeping results proprietary. That stance made FAIR a counterweight to the more closed approach of some rivals, and it produced widely used open tools and models, most prominently the PyTorch deep-learning framework and, later, the open-weight Llama family.
The founding of FAIR was part of a wave around 2013 to 2014 in which big technology companies raced to build serious AI research organizations, alongside Google Brain, the acquisition of DeepMind, and similar moves elsewhere. It signaled that deep learning had graduated from an academic curiosity into a strategic corporate priority.
Why business readers should care: FAIR showed that a consumer technology company could run a top-tier open research lab, and the open models and frameworks it released, especially PyTorch and Llama, became foundations that countless other businesses now build on.