
TL;DR
- Meta confirms Shengjia Zhao as Chief Scientist of Meta Superintelligence Labs (MSL).
- Zhao played a critical role in OpenAI’s development of ChatGPT, GPT-4, and o1.
- Zhao joins under the leadership of Alexandr Wang, former CEO of Scale AI.
- Meta has hired top talent from OpenAI, DeepMind, Apple, and Anthropic.
- MSL will gain access to Prometheus, a 1-gigawatt AI training cluster, by 2026.
- Meta now has two Chief AI Scientists: Zhao (MSL) and Yann LeCun (FAIR).
Meta Formalizes Zhao’s Role at MSL
Meta CEO Mark Zuckerberg officially named former OpenAI researcher Shengjia Zhao as Chief Scientist of Meta Superintelligence Labs (MSL). The announcement was made via Threads on Friday, emphasizing Zhao’s critical role in leading Meta’s push into frontier AI research.
Zhao was one of the foundational minds behind ChatGPT, GPT-4, and the first AI reasoning model o1 at OpenAI. His appointment reflects Meta’s intent to directly challenge OpenAI and Google DeepMind in the next phase of AI development.
Strategic Addition Under Alexandr Wang’s Leadership
Zhao will work alongside Alexandr Wang, the newly appointed head of MSL and former CEO of Scale AI. While Wang brings deep operational experience, his non-research background led some to question his fit as leader of an AI R&D lab.
By naming Zhao — a recognized scientist in AI reasoning architectures — as Chief Scientist, Meta shores up its technical leadership. According to Zuckerberg’s Threads post, Zhao has co-led the lab from its inception and is the driving force behind its current R&D direction.
Meta Recruits Elite AI Talent from OpenAI and Beyond
Zhao’s appointment is part of a broader, aggressive recruitment strategy led by Zuckerberg himself. According to The Information, Meta has secured multiple high-profile AI researchers, including:
- Jiahui Yu, Shuchao Bi, and Hongyu Ren, all formerly at OpenAI
- Trapit Bansal, who worked directly with Zhao on reasoning models
- Multiple researchers from OpenAI’s Zurich office, specializing in multimodal systems
Additionally, internal talent from Meta’s FAIR and Generative AI units is being funneled into MSL.
Infrastructure Power: Prometheus AI Cluster
To support MSL’s mission, Meta is scaling up its cloud infrastructure investment. The company’s Prometheus training cluster, currently under development in Ohio, will go online by 2026. Once operational, it will deliver 1 gigawatt of compute power — enough to serve more than 750,000 homes.
This would make Meta one of the first tech companies to operate such a massive AI training environment, allowing Zhao and team to run large-scale frontier model experiments.
AI Leadership & Compute at a Glance
Metric | Details |
MSL Chief Scientist | Shengjia Zhao |
Prometheus Compute Cluster | 1 gigawatt by 2026 |
Leadership Background | Zhao (AI Research), Wang (Scale AI Ops) |
Major Recruits | From OpenAI, Anthropic, Apple, Google DeepMind |
Meta’s Existing AI Units | FAIR (long-term research), Generative AI (products), MSL (frontier models) |
Meta’s Dual-Scientist Strategy: Zhao and LeCun
Zhao now joins Yann LeCun, head of Meta’s FAIR lab, as a second chief scientist within the company. LeCun focuses on long-term research in fundamental AI science, while Zhao’s mandate is to deliver competitive frontier AI models within a shorter timeframe.
The delineation between FAIR, Generative AI, and MSL is still evolving. But the addition of Zhao ensures Meta has deep academic leadership to rival both OpenAI’s Ilya Sutskever and Google’s Demis Hassabis.
Meta’s Superintelligence Vision Accelerates
Zuckerberg’s approach to building MSL has involved direct outreach to top researchers, personal invitations to Lake Tahoe, and aggressive compensation. According to insiders, some offers are “exploding,” with expiration deadlines measured in days.
With Zhao at the helm, MSL is expected to prioritize reasoning, alignment, and multimodal AI — positioning itself directly against OpenAI’s o1 model and future iterations of GPT.
Final Thoughts
Meta’s appointment of Shengjia Zhao is more than a promotion — it’s a signal of intent. With two world-class scientists leading distinct branches of AI development, Meta now sits firmly among the elite in the superintelligence race. As MSL scales its compute and brainpower, the next wave of AI innovation might well come from Zhao’s lab — not just from Silicon Valley’s usual suspects.