Photo: Business Insider
Meta CEO Mark Zuckerberg has officially announced the creation of Meta Superintelligence Labs (MSL) — a bold initiative aimed at positioning the company at the forefront of artificial intelligence development.
In an internal memo obtained by CNBC, Zuckerberg outlined the strategic purpose of MSL: to consolidate and supercharge Meta’s foundation model efforts, including its widely used Llama open-source AI models, core research initiatives, and product development efforts powered by generative AI.
The lab will be led by newly recruited heavyweights in the AI world, including former Scale AI CEO Alexandr Wang, now serving as Meta’s Chief AI Officer, and former GitHub CEO Nat Friedman, along with tech entrepreneur Daniel Gross.
Zuckerberg revealed that MSL will be responsible for advancing Meta’s current and next-generation AI models. Work is already underway on Llama 4.1 and 4.2, which power Meta AI — the company’s suite of intelligent agents used by over 1 billion monthly active users across Facebook, Instagram, WhatsApp, and other platforms.
In parallel, Meta is initiating research on what Zuckerberg referred to as its "frontier" generation of AI models, with the goal of reaching superintelligence-level capabilities within the next year.
“I’ve spent the past few months meeting top minds across Meta, leading AI labs, and fast-moving startups to assemble the founding group for this tightly focused and talent-dense effort,” Zuckerberg wrote. “We’re still building this team and will continue adding members from within the company’s broader AI organization.”
Zuckerberg emphasized that Meta holds unique advantages in the race toward superintelligent systems:
“We’re committed to building personal superintelligence that benefits everyone — not just a few enterprises or governments,” Zuckerberg said.
Meta’s aggressive entry into advanced AI is intensifying the war for top talent. According to reports, Meta offered signing bonuses as high as $100 million to recruit researchers from OpenAI. While OpenAI CEO Sam Altman acknowledged the recruiting pressure, other companies like Microsoft are also reportedly countering Meta’s offers.
Earlier this year, Meta attempted to acquire Safe Superintelligence, an AI safety-focused startup co-founded by Ilya Sutskever, a former OpenAI executive. The bid was unsuccessful, but Meta ultimately brought on board Friedman and Gross, both formerly associated with the venture.
Meta has also committed a $14.3 billion investment into AI infrastructure, including its deal to bring Wang and his team from Scale AI on board.
With MSL, Meta is betting big on the future of open-source AI and distributed superintelligence. The company’s roadmap includes:
Meta has stated its ambition to democratize AI — not only through open models, but also by embedding intelligent agents into every layer of its ecosystem. This includes chatbots, smart assistants, AR devices, and more.
“This is just the beginning,” Zuckerberg concluded. “We have even more incredible talent joining the effort in the coming weeks. I’m excited to dive in and build what’s next.”
As the AI race accelerates, Meta’s Superintelligence Labs could redefine the global competition — not just among Big Tech companies, but across how artificial intelligence reshapes economies, societies, and the human experience.