“Today, as the world undergoes significant change, as the race for AI accelerates, and as Meta prepares for its next chapter, it is time to create space for others to pursue the work,” Meta’s outgoing head of artificial intelligence research, Joelle Pineau, wrote on LinkedIn. Pineau’s departure is a turning point for Meta, with the company struggling to keep up in an increasingly competitive environment for AI as it shells out billions of dollars to remain in contention. She leaves on May 30th, and there is an empty seat in a position that has been spearheading Meta’s AI efforts.
Pineau, who was at Meta since 2017, was the head of Fundamental AI Research (FAIR), a group of about 1,000 researchers at 10 sites worldwide. During her time at FAIR, the team wrote about work on voice translation, understanding images, and creating Meta’s open-source Llama large language model. FAIR’s use of an open-source strategy has been welcomed by technologists for the way it is able to democratize the creation of AI, even if it has baffled Wall Street with its unconventional means of making money. Meta CEO Mark Zuckerberg has made AI central to the company’s future plans, investing up to $65 billion in AI-related projects this year alone.
Among these initiatives, Llama stands out as a key component of Meta’s strategy to rival proprietary models like OpenAI’s GPT and Google’s PaLM 2. Zuckerberg envisions Llama as the industry standard, powering AI chatbots across Meta’s platforms and potentially reaching over a billion users annually. As of December, Meta AI had already reached 600 million monthly users, indicating its increasing prominence. But Pineau’s exit can become harder to enact such plans. Industry experts observe that leadership changes in AI research can play a big role in determining the ability of a firm to innovate and provide. The firm has yet to appoint a successor, with concerns over how it will manage to keep up in its race in AI.
Pineau’s departure timing is also consistent with standard Meta leadership change such as vice presidency of Asia-Pacific operations, and departure managing director Kate Hamill who previously led North American retail and e-commerce business. Moves of these top-management exits followed as Meta battles to lose customers to competing sets of groups such as OpenAI, Anthropic, and Elon Musk-owned xAI companies in their race to pioneer constructing next-page AI innovations.
Pineau’s contributions to Meta also have a lot to do with her passion for open science and research. In an interview with the Financial Times, she had indicated why she had joined Meta by saying,”It was pretty obvious that a lot of the biggest innovation in AI was going to happen in industry.” She had also commended Meta on its commitment to open-source principles, which she had seen as one of the tech industry’s best.
Meta’s open source approach, as seen in Llama, has generated headlines.
Meta made the first Llama model available in February, with between 7 billion and 65 billion parameters. The models are designed to be capable and economical, on Meta’s Research SuperCluster supercomputer with 16,000 Nvidia A100 GPUs. The just released Llama 2, which was released in July, has already been downloaded over 30 million times onto platforms such as Hugging Face, a clear reflection of how popular it is among researchers and developers. This notwithstanding, Llama’s open-source model has not been controversy-free. It has drawn criticism from some as being contradictory to the open innovation spirit because Meta’s terms of licensing prohibit use for products with over 700 million monthly active users.It looks like Meta wants all the benefits of open source for their business while keeping the competition away, is the view of Umesh Padval, a venture partner at Thomvest Ventures.
As Meta continues, it also has the twin mission of keeping its technological advantage but having to deal with transparency and governance issues. The company spokesperson expressed its commitment to developing AI responsibly when it said, “We believe in open innovation, and we do not want to place undue restrictions on how others can use our model. However, we do want people to use it responsibly.”
Pineau’s exit is a reminder of the human factor involved in AI development. While Meta’s investments and infrastructure are essential, the vision and ability of its leadership are also key in determining the future of the company. While users are working toward creating an AI assistant with artificial general intelligence and growing Llama’s impact, succession planning for choosing Pineau’s successor must be a top priority of Meta’s AI goals.
The stakes are high, not only for Meta but also for the entire tech world.
Whereas Microsoft, Amazon, and Google are all set to invest $525 billion a year in AI in 2032, the competition game is becoming increasingly challenging. For Meta, the response will be time itself: Can it maintain its innovation lead despite having to ride through the turbulence of leadership change and open-source innovation? Only time itself can tell.
