Meta boosts AI spending to $145 billion despite user decline

Meta is spending more on AI even as its daily audience showed a rare sequential dip, a combination that sharpened the tension between its social-media core and its bet on a much larger computing future. The company lifted its 2026 capital expenditure outlook to $125 billion to $145 billion, up from a prior range of $115 billion to $135 billion, while reporting 3.56 billion average daily active people in March 2026.

That user figure was still up 4% from a year earlier, but it fell from the prior quarter. Meta said the decline was driven by internet disruptions in Iran and restrictions on access to WhatsApp in Russia. Wall Street had expected about 3.62 billion daily active people, making the print a miss against estimates even as the company insisted the underlying business remained strong.

The numbers on the ad business were much harder to dismiss. First-quarter 2026 revenue rose 33% from a year earlier to $56.31 billion, helped by a 19% increase in ad impressions and a 12% rise in average price per ad. Meta’s first-quarter costs and expenses were $33.44 billion, and capital expenditures including principal payments on finance leases reached $19.84 billion.

Mark Zuckerberg framed the quarter as a turning point. He called it a “milestone quarter” and said Meta had released its first model from Meta Superintelligence Labs, adding that the company was on track to deliver “personal superintelligence to billions of people.” The pitch underscored how aggressively Meta is tying its future to AI infrastructure, models and talent.

That spending push has already reshaped the company’s balance sheet and its narrative. Meta ended 2025 with more than 3.5 billion people using at least one of its apps every day, including more than 2 billion daily actives each on Facebook and WhatsApp and just shy of 2 billion on Instagram. The latest numbers suggest that scale is still immense, but no longer immune to regional disruptions or signs of saturation.

Meta said higher component pricing and additional data center costs helped drive the higher capex outlook, a reminder that the AI race is becoming a capital-intensity contest as much as a software contest. Shares fell about 7% in extended trading after the results, as investors weighed a softer user print against a far heavier investment bill and a company that is betting AI can offset any weakness in its core social business.