Meta's AI Superintelligence Push Sparks Internal Talent Friction
Meta’s ambitious pursuit of artificial superintelligence (ASI) is reportedly creating significant internal friction within its sprawling AI operations, sparking concerns among existing researchers and prompting threats of desertion. While the tech giant pours billions into acquiring elite talent and infrastructure to lead the global AI race, its aggressive strategies are simultaneously alienating some of its long-standing experts.
The core of the tension stems from Meta’s dramatic pivot towards AI, a shift that has seen the company move resources and focus from its earlier metaverse ambitions. Mark Zuckerberg has articulated a bold vision for “personal superintelligence” for everyone, establishing a dedicated 50-person Meta Superintelligence Lab (MSL) to spearhead this effort. However, this intense focus and the accompanying “cash-driven” talent acquisition strategy—with reports of record-breaking pay packages, some reaching $100 million or even $300 million over four years, and hefty signing bonuses—are causing unease among the existing workforce.
Sources indicate that Meta’s aggressive recruitment, which includes maintaining a “literal list” of desired AI specialists poached from rivals like OpenAI, Google, Apple, and Anthropic, can make existing researchers feel undervalued or sidelined. While the allure of extraordinary financial rewards is undeniable for new recruits, some established employees are reportedly skeptical of Meta’s tactics, expressing concerns about potential burnout, shifts in company culture, and ethical alignment with the superintelligence push. Companies like Microsoft and Anthropic, which emphasize creative freedom, flatter hierarchies, and a strong focus on AI safety and ethics, often appeal more to those prioritizing purpose and work environment over purely financial incentives. The recent departure of Joelle Pineau, Meta’s Vice President for AI research, in May 2025, who cited the accelerating AI race and Meta’s “next chapter” as reasons to “create space for others,” further underscores the dynamic internal shifts occurring within the company.
This internal discord is unfolding amidst a broader, fiercely competitive AI talent war across the industry. With Meta committing substantial financial resources—projected capital expenditures between $66 billion and $72 billion in 2025 for AI infrastructure, including a massive deployment of 600,000 GPUs—the company is undeniably making a colossal bet on its AI future. This spending, however, comes even as Meta, like other tech giants, has undertaken significant layoffs, creating a paradoxical environment where vast sums are spent on new elite talent while other parts of the workforce face uncertainty. The impact of Meta’s poaching is already evident, with rivals reporting talent attrition and project delays as their top researchers are lured away.
Ultimately, Meta faces the delicate challenge of balancing its audacious superintelligence goals with maintaining morale and retaining the expertise of its existing, diverse AI teams. The race for AI dominance is not merely about acquiring the brightest new minds, but also about fostering an inclusive and supportive environment where all contributors feel valued and aligned with the company’s evolving strategic direction, especially as ethical considerations around superintelligent AI become increasingly prominent.