Meta has tweaked its full-year capital expenditure guidance for 2025, tightening the range to $66 billion to $72 billion, up slightly from the previous lower bound of $64 billion. It’s a modest adjustment on paper, but it signals something bigger, a company doubling down on artificial intelligence, even as it braces for rising costs in the years ahead.
The real spike, however, is expected in 2026. And it won’t be hardware or data centres leading the charge. According to Meta’s finance chief Susan Li, it’s going to be people. Specifically, the highly sought-after kind with AI credentials.
AI Talent: Expensive, Essential
“Within AI, we’ve had a particular emphasis on recruiting leading talent within the industry, as we build out Meta Superintelligence Labs to accelerate our AI model development and product initiatives,” Li said during Meta’s July 30 earnings call.
Meta isn’t just dabbling in AI, it’s pouring serious money into it. And as the hiring wave continues through 2025, Li indicated that 2026 will bear the full financial weight of that decision. She stopped short of offering a hard number, noting that it’s “still very early in planning for next year,” but confirmed that cost growth in 2026 will outpace the 20 to 24 per cent increase expected this year.
That increase, she added, will be “driven by a full year of compensation expense for the AI talent the company is hiring this year.”
Building The Brain Trust
Meta’s strategy hinges on scale and speed. By establishing its Superintelligence Labs, the company is attempting to leapfrog competition in the generative AI space. That means not only developing more advanced models but also integrating them across platforms like Facebook, Instagram, and WhatsApp.
Li also reiterated Meta’s continued focus on hiring in “priority areas such as AI, infrastructure and reality labs.” In other words, the company isn’t taking its foot off the gas.
Betting Big On AI Arms Race
This isn’t just another budget reshuffle. Meta’s latest projections suggest a long-term view, one where it’s willing to endure a spike in operational costs to secure a competitive edge. The tech giant is making it clear: to win the AI race, you need more than servers, you need the smartest minds, and they don’t come cheap.
Whether this massive investment in talent and labs translates into market-defining products is still uncertain. But for now, Meta seems prepared to pay the price.
