Meta's latest guidance is a blunt reminder that the AI buildout is still accelerating even when investors flinch at the bill. According to a Reuters report published April 29, the company lifted its 2026 capital expenditure range to $125 billion-$145 billion from a prior $115 billion-$135 billion. The stock fell more than 6% in extended trading. That is the market's way of saying enthusiasm for AI remains high, but patience on payback is getting thinner.

The interesting part is that Meta did not pair the higher capital bill with an obviously weaker operating picture. Reuters says first-quarter revenue reached $56.31 billion, above analysts' $55.45 billion estimate, and daily active people were 3.56 billion, up 4% from a year earlier. On a narrower reading, the core ad engine is still doing the job of funding the AI transition. The tension is that investors now want proof that ever-larger infrastructure checks will create more than a stronger narrative.

Meta is also trying to squeeze savings and output from the workforce at the same time. Reuters reports planned layoffs continue into the second half, while the company reshapes teams around AI-heavy workflows. The same article notes new internal tracking software on U.S.-based employees' computers as part of a broader push to train AI agents that can handle work tasks more autonomously. That makes the company a useful case study in how the AI buildout is changing not just capex lines, but management culture and labor assumptions inside large tech firms.

There is outside pressure, too. Meta is dealing with youth-safety litigation and growing regulatory scrutiny even as it spends harder on AI. That combination matters. The company is trying to finance a more compute-intensive future from an advertising machine that still throws off cash, while absorbing legal risk and demanding more from fewer people. The new capex range says Meta believes the cost of underbuilding for AI is now worse than the cost of spooking Wall Street for another quarter.

#hyperscalers

Meta lifts AI capex to $145B, and investors still want the return