Meta CEO launches 50-person AI group to regain edge after Llama 4 setbacks.
Meta’s new AI unit is based at the company’s Menlo Park headquarters.
- Meta CEO Mark Zuckerberg is assembling a 50-person “superintelligence group” to focus on artificial general intelligence (AGI).
- The team, made up of top AI researchers and engineers, will report directly to Zuckerberg and work near his office.
- Meta has increased its AI-related spending to as much as $72 billion amid concerns over the performance of its latest AI model, Llama 4.
Zuckerberg Assembles New AI Power Team
Meta Platforms CEO Mark Zuckerberg is spearheading the creation of a high-level team of 50 artificial intelligence experts to accelerate the company’s push toward artificial general intelligence (AGI)—a goal defined by the development of AI systems capable of human-level reasoning and learning. The new team is referred to internally as the “superintelligence group.”
Strategic Location and Leadership
This elite group of AI engineers and researchers will be located near Zuckerberg’s office at Meta’s Menlo Park headquarters in California. His direct involvement and proximity to the group underscore the personal significance he places on the project and Meta’s broader AI ambitions.
Llama 4 Frustration Sparks Reboot
According to individuals familiar with the effort, Zuckerberg’s intensified focus comes after dissatisfaction with the reception and performance of Llama 4, Meta’s latest large language model. The company recently delayed the release of a new “Behemoth” version of the model due to insufficient improvements. This has prompted Zuckerberg to take a more hands-on role in AI team recruitment and development.
Meta Boosts Spending to Regain AI Leadership
To support its AGI goals, Meta announced it will raise its capital expenditures for 2025 to between $64 billion and $72 billion—an increase designed to scale up AI infrastructure and research capacity. As of Tuesday morning, Meta shares had risen roughly 1% in premarket trading, continuing a nearly 20% gain so far this year.
Background: Meta’s AI Journey and Setbacks
Meta has long positioned itself as a major player in AI, but recent setbacks, including mixed reviews of its Llama 4 model, have called its leadership into question. Despite its earlier innovations, the company has faced criticism over the pace and quality of its advancements compared to rivals.
Artificial general intelligence represents the next frontier in AI, with the potential to transform industries through machines that can learn and reason like humans. By assembling a dedicated “superintelligence” team and dramatically increasing investment, Zuckerberg is betting on a rapid rebound—and a leadership role—in the race for AGI.
Final Thoughts
Zuckerberg’s creation of a “superintelligence group” marks a pivotal shift in Meta’s AI strategy, reflecting both urgency and ambition. With increased investment and direct executive oversight, Meta is determined to regain its footing in the rapidly evolving AI race. For more on AI’s role in media narratives, see this analysis of AI and media bias. To explore Meta’s broader AI roadmap, check out key takeaways from Zuckerberg’s 2024 Meta Connect keynote.
