Big Tech’s big AI buildout will not end until someone stops it

Amazon, Microsoft, Alphabet, and Meta plan to spend at least $720 billion on capital expenditures this year, mostly on AI infrastructure, adding to over $1 trillion in future commitments. This spending spooked Wall Street, prompting a tech selloff, but stocks rebounded as investors focused on midterm elections that could lead to data center regulations. Despite strong revenue growth—Amazon's fastest in five years, Meta's ad surge, and Google Cloud's doubling—profits from AI remain elusive. Alphabet's stock suffered its worst one-day drop (13%) after acknowledging that AI Mode queries have lower monetization, hurting Search ad growth (12% vs. 17% last year). CFO Anat Ashkenazi said spending will increase "significantly" in 2027, while Amazon CEO Andy Jassy called the spending a "once in a lifetime opportunity" and plans to spend "meaningfully more" than $220 billion next year. Meta expects to spend $135–145 billion this year and increase next year, with CEO Mark Zuckerberg noting that agentic AI profitability depends on reducing compute costs. Anthropic's Q3 revenue hit $17 billion but compute costs were $17.5 billion, with nearly $900 billion in future commitments and plans to build its own data centers. Google sells TPUs to Anthropic, raising questions about capacity constraints. Microsoft, the only company to avoid Wall Street's ire with lower-than-expected capex due to "third party capacity delays," subsequently announced a $25 billion bond sale. Big Tech stocks recovered, possibly on hopes that midterm election results will force spending slowdowns, giving more time to achieve AI profitability. Notably, midterm elections were not mentioned on any earnings calls.

(Source：The Verge)