Meta Platforms, the company formerly known as Facebook, is facing escalating internal friction as it doubles down on artificial intelligence amid intensifying competition with rivals OpenAI, Google, and Anthropic. The tech giant’s push for AI supremacy now involves multibillion-dollar investments, bold reorganizations, and headline-grabbing hires—moves that, while strategically ambitious, are sowing discord within the company’s vaunted research ranks.

According to insiders and multiple recently published reports, the arrival of high-profile recruits such as Alexandr Wang, newly installed as Chief AI Officer following a $14.3 billion “acquihire” of his company Scale AI, and Shengjia Zhao, the OpenAI veteran who helped co-create ChatGPT and was appointed Chief AI Scientist in July, has sparked a power struggle with Meta’s established engineering and research teams.

A New Leadership Team Changes Meta’s AI Course

The contrast between the new AI leadership’s vision and the company’s traditional approach is striking. Wang’s “Superintelligence” team advocates for a more secretive, closed approach to AI model development—sharply diverging from Meta’s long-held philosophy of open-sourcing its models, which has allowed communities and external developers to build upon innovations such as the Llama model series.

Shengjia Zhao, meanwhile, is conducting intensive performance assessments and role reviews of Meta’s longstanding engineers, prompting a sense of scrutiny—or outright displacement—among much of the old guard. Researchers have reportedly queued outside Zhao’s office for interviews that will determine their place in the company’s restructured hierarchy.

“There is a palpable tension as founding researchers and engineers face evaluation from new leaders brought in with sky-high expectations and compensation,” said a senior Meta AI engineer who requested anonymity. “The stakes are enormous—not just for individuals, but for the future direction of one of the world’s largest tech platforms.”

Massive Investment and Strategic Shifts

In the 2025 fiscal year, Meta is expected to invest up to $72 billion in capital expenditures, much of which is earmarked for AI R&D and infrastructure. This comes on top of the $14.3 billion paid for a 49% stake in Scale AI—a bold bid to secure talent and technology capable of outpacing OpenAI and Google DeepMind in the race to develop ever more powerful AI models and foundation technologies.

During Meta’s Q2 2025 earnings call, CEO Mark Zuckerberg characterized this moment as “a new era of individual empowerment” while assuring investors that aggressive AI investments are vital to retaining relevance as the industry transitions from social media to AI-driven products. These investments have elevated Meta into a power center for AI recruitment, with Silicon Valley seeing a bidding war for top researchers and engineers.

“AI is the engine of future growth and engagement, not only for Meta but for every tech platform that intends to be relevant in the next decade,” said industry analyst Rachel Chou of EmergeTech Capital. “The flip side is that integrating such an influx of high-profile talent without losing institutional knowledge is a near-impossible balancing act.”

Old Guard vs. New: Disputed Philosophy and a Talent Exodus

The leadership changes coincide with an exodus of veteran talent. Notably, Joelle Pineau—a prominent figure in the company’s AI ethics and research agenda—departed and joined Cohere, a leading competitor in natural language AI. Angela Fan, known for her pivotal role in developing the open-source Llama models, left for OpenAI, one of Meta’s key rivals. Loredana Crisan, VP of generative AI, is migrating to Figma as its chief design officer.

The rift is not just personal; it’s philosophical. Wang’s push to abandon open-sourcing marks a significant risk, as Meta’s dominance in developer communities has been predicated on transparency and cooperation. The decision to halt Meta’s previous “Behemoth” frontier model after poor performance tests underscores the leadership’s willingness to make sweeping changes when results fall short of expectations.

To compound matters, the company is reportedly considering downsizing its ballooning AI division, which now numbers in the thousands. Potential layoffs and internal transfers are on the table in an effort to streamline operations and focus resources on the new AI superintelligence strategy.

Competitiveness and Controversy: The Road Ahead

Beyond internal realignment, Meta is exploring contracting or licensing third-party AI models—a dramatic departure from its self-reliant past. Industry insiders suggest that partnerships, whether with open-source communities or closed-source tech leaders, could help Meta turbocharge its AI efforts in the face of Google, Anthropic, and OpenAI’s rapid advances.

Public reaction to the shakeup has been mixed, with Meta spokesperson Andy Stone ridiculing media coverage as “sensationalist” and insisting that the AI restructuring is portrayed out of context. Nevertheless, industry observers agree that Meta is mounting its boldest gamble yet as the existential AI race fundamentally reshapes the hierarchy of Silicon Valley’s elite.

Competition for superintelligence talent and computing resources is so fierce that even chief executives are directly involved in recruitment, offering unprecedented compensation and benefits as major tech players seek an edge. Meta’s moves are mirrored by rival strategies: Microsoft’s continued backing of OpenAI, Google’s integration of Gemini into its entire product suite, and Amazon’s recent $10 billion investment in AI startup Anthropic have all intensified the race.

Looking Forward: Can Meta Stay Ahead?

As the AI revolution evolves, all eyes are on whether Zuckerberg’s vision can steer Meta safely through these stormy waters. The outcome will have lasting consequences for Meta’s product portfolio—from social networking to mixed reality and workplace applications—as well as the global tech ecosystem’s broader approach to AI development, ethics, and openness.

Meta’s story is one of ambition and risk: consolidation of world-class talent, the re-imagining of established models, and the renewal or loss of core company culture. Whether the new AI leadership’s closed approach will pay off as AI models reshape society, or whether internal rifts and lost talent hamper progress, remains to be seen. For now, Meta is unmistakably at the center of Silicon Valley’s AI arms race—and the whole industry is watching.