Meta Platforms Inc. (NASDAQ: META), the tech giant behind Facebook, Instagram, and WhatsApp, is making a bold strategic bet on the future of artificial intelligence (AI). Under CEO Mark Zuckerberg’s direction, Meta is pivoting its AI research toward small, “talent-dense” teams and launching secretive new divisions with an ambitious goal: to push the boundaries of superintelligent AI.

The centerpiece of this strategy is a newly formed, elite group known as TBD Lab, spearheaded by Alexandr Wang, founder of Scale AI. The high-profile recruitment of Wang from his unicorn startup, alongside former GitHub CEO and tech investor Nat Friedman, marks one of the most aggressive talent plays in Silicon Valley’s ongoing AI arms race. Their mission is clear: accelerate Meta’s progress toward superintelligence by emulating the energy and agility of startups.

Meta’s Strategic Shift: Superintelligence Labs and Lean Innovation

In recent months, Meta has invested billions in restructuring its AI organization. The company’s $14 billion stake in Scale AI, as reported by Reuters, cemented its commitment to cutting-edge AI infrastructure and data management. The reorganization, according to internal memos and Business Insider reports, divides Meta’s AI division into four specialist units aimed at high-impact frontier research.

On the company’s latest earnings call, Zuckerberg emphasized his conviction that “small, talent-dense teams” outperform larger, more bureaucratic groups in driving research breakthroughs. “For leading research on superintelligence, you want the smallest group that can hold the whole thing in their head,” Zuckerberg stated, underscoring Meta’s cultural pivot from scale to agility.

This approach is inspired by successful precedents in the industry. Google’s revolutionary “Attention Is All You Need” paper, which laid the groundwork for the transformer models powering today’s AI, was authored by just eight engineers. Similarly, rising AI startups like Hightouch, valued at $1.2 billion, achieve outsized results with small, focused engineering teams.

The New Faces: Alexandr Wang and Nat Friedman

Meta’s recruitment of Alexandr Wang, acclaimed for building Scale AI into a multibillion-dollar data-labeling powerhouse, sends a clear message to the industry: Meta is doubling down on high-caliber leadership. Wang’s role now extends beyond infrastructure, shaping Meta’s broader superintelligence strategy and overseeing cutting-edge research teams.

Nat Friedman, a respected investor and former CEO of GitHub (acquired by Microsoft for $7.5 billion), joins as head of Products and Applied Research. Friedman is well-known for championing small teams and developer agility, having publicly advocated that “smaller teams are better—faster decisions, fewer meetings, more fun.” Their combined experience signals a new era at Meta, where product focus and technical excellence are prioritized above legacy approaches.

A Talent War Unlike Any Other

Meta’s aggressive ramp-up of its superintelligence push has intensified the AI talent war. In the past year, Meta has reportedly offered nine-figure compensation packages to lure engineering talent from rivals like OpenAI, Google DeepMind, Anthropic, and other top AI labs. The Wall Street Journal notes that these unprecedented offers reflect a belief that “a handful of world-class researchers can have outsized impact.”

While these high-stakes moves promise faster innovation, they risk internal friction. Several long-serving Meta engineers have expressed dissatisfaction with lavishly funded new units, with reports of morale issues and threatened resignations. “It can be disruptive, but you simply can’t compete for superintelligence using old playbooks,” Wang wrote in an internal memo. This realignment reflects broader Silicon Valley trends as well, where engineers at Hightouch and other smaller AI startups are empowered to prioritize projects and minimize managerial overhead.

“You don’t need too many—just a few smart, cream-of-the-crop people to have major breakthroughs,” said Meta AI engineer Yangshun Tay, echoing a sentiment now widely adopted across the industry.

Can the Startup Model Scale at Big Tech?

Despite historical successes of small teams, scaling this model within massive companies like Meta is fraught with complexity. Elliott Parker, CEO of Alloy Partners, notes that “small teams inside conglomerates can drive useful products, but rarely transform their parent organization wholesale.” Meta has already reorganized its AI units multiple times in 2025, dissolving overlapping divisions to streamline operations—a move insiders likened to removing ‘slime mold’ for renewed focus.

Nevertheless, Meta’s leadership remains bullish. Rapid launches like the Llama 3 large language model, partnerships with OpenAI’s competitors, and record investments in AI research infrastructure demonstrate Meta’s resolve to be at the industry’s bleeding edge. Global AI spending is estimated to reach nearly $300 billion by 2026 (IDC), intensifying the need for innovation at scale.

Meta’s superintelligence bet faces not just technical hurdles but also challenges in retaining top talent, integrating new cultural norms, and keeping pace with nimble startups. However, if successful, this hybrid “startup inside a giant” model could redefine how big tech spearheads the next era of artificial general intelligence.

Industry Implications and What’s Next

The race for AI superintelligence has far-reaching implications across the tech landscape. Meta’s approach signals a willingness to disrupt itself in pursuit of leadership—potentially influencing how Google, Microsoft, and Amazon structure their own AI teams. The induction of top-tier talent like Wang and Friedman could attract further elite researchers to Meta, emboldening smaller teams across the industry to challenge established norms.

As the balance of power in AI shifts from size to agility, Meta faces a defining test: can it translate the energy and risk-taking of a startup into breakthroughs within a $1.2 trillion corporation? The answers over the coming year are likely to shape not just Meta’s fate, but the next chapter of global AI innovation as a whole.