Mark Zuckerberg Takes Control!
In the face of disappointing results from its recent large language models (LLMs), Meta’s CEO, Mark Zuckerberg, is stepping up efforts to reposition his company at the forefront of the global race toward Artificial General Intelligence (AGI). According to Bloomberg reports, Zuckerberg has quietly initiated the formation of a new elite team dedicated to “superintelligence,” with the explicit goal of surpassing competitors such as OpenAI, Google, and Anthropic.
He is prepared to invest heavily in this project, personally overseeing recruitment for a specialized team of approximately 50 top-tier experts. This initiative signals a significant strategic shift, emphasizing Meta’s ambition to lead in the next generation of AI technology.
Investing Billions in American Start-up Scale AI
Central to Zuckerberg’s strategy is a multi-billion dollar investment in the American startup Scale AI, a company renowned for providing data services crucial to training advanced AI models. Founded by Alexandr Wang, Scale AI specializes in preparing datasets for machine learning applications. Sources indicate that Wang is expected to join Meta’s “superintelligence” team once negotiations conclude. This investment could exceed $10 billion, reinforcing Meta’s serious commitment to advancing its AI capabilities.
Integrating AGI Across Meta’s Ecosystem
This maneuver follows a period of internal frustration at Meta. The company’s latest attempt at an LLM, called Llama 4, released in April, has failed to meet expectations—neither the internal teams nor outside developers have been satisfied with its performance. Moreover, a project dubbed “Behemoth,” which was supposed to mark a major breakthrough, was shelved. The leadership deemed that the model did not outperform its predecessors sufficiently to warrant release, as reported by The Wall Street Journal.
In response to these setbacks, Zuckerberg has become directly involved in rekindling Meta’s AI ambitions. He formed a WhatsApp group called “Recruiting Party,” consisting of his key lieutenants, to identify and prioritize top talent for recruitment. Zuckerberg is said to personally manage early communications and closely monitor every phase of the hiring process, according to Bloomberg.
His vision is to embed AGI into all of Meta’s products—spanning social media platforms, the “smart” Ray-Ban glasses, and conversational assistants. Zuckerberg asserts that Meta’s robust advertising business model will finance these ambitious infrastructure projects, including the development of a multi-gigawatt data center, claimed to be among the most powerful in the world.
The Future of Meta’s “Superintelligence Labs” and Its Possible Team Dynamics
It remains to be seen how this new group—internally dubbed “Superintelligence Labs”—will interact with existing AI teams at Meta. Notably, Yann LeCun, Meta’s Chief Scientist for AI, who played a crucial role in founding Facebook AI Research (FAIR), based in Paris, is expected to be involved. Known to be close to Zuckerberg and often skeptical of the notion that AGI is imminent or solely based on language models, LeCun’s perspectives may influence how the new division develops.
Much like other tech giants investing heavily in AI, Meta’s strategic push hints at a broader industry trend: a relentless race to develop powerful, practical, and potentially transformative artificial intelligence technologies. Zuckerberg’s hands-on approach, combined with significant financial backing, underscores Meta’s desire to not only catch up but to lead the next frontier in artificial intelligence research and application.
