Meta Superintelligence Labs (MSL) has been officially established within Meta, tasked with developing next-generation artificial intelligence (AI) models for Zuckerberg's company. The establishment of MSL was announced by Zuckerberg in a memo to employees, according to a Wired report.
The team that builds AI models, products, and projects at Fundamental Artificial Intelligence Research (FAIR) is placed under the MSL umbrella. As we previously reported, former Scale AI CEO Alexandr Wang will lead MSL. In the same memo, Zuckerberg announced several new employees joining MSL from rival AI companies.
Among them are former GitHub CEO Nat Friedman and former CEO and co-founder of Safe Superintelligence Daniel Gross. Gross joined Meta after Zuckerberg failed to buy Safe Superintelligence, which was also founded by Ilya Sutskever, who was formerly the head of research at OpenAI.
In addition, several researchers from OpenAI were successfully poached, including Shengjia Zhao, who is the co-creator of ChatGPT. Also successfully poached from OpenAI were the creators of the o3/o4-mini, GPT-4o, GPT-4.1, GPT-4.5, 4o-imagegen, GPT-4o voice mode and o4-mini models. The heads of Gemini, Waymo and Deepmind development were also successfully poached from Google by MSL.
Sam Altman previously said that OpenAI employees rejected the $100 million bounty offered by Meta. But now it is clear that key researchers took Zuckerberg's offer, forcing OpenAI to recalibrate its staff compensation structure.