Alphabet Raises the Stakes on AI, Setting a New Benchmark for Infrastructure Spending

Alphabet delivered stronger-than-expected fourth-quarter results, beating forecasts on revenue, earnings per share and cloud performance. Yet the positive momentum was overshadowed by the company’s latest guidance on artificial intelligence infrastructure spending, which reset expectations for the entire sector.

The company said it expects capital expenditures in 2026 to reach between $175 billion and $185 billion. At the high end, that would represent more than double its 2025 spending and place Alphabet well ahead of other hyperscalers in absolute investment. The projection follows earlier signals that spending would rise sharply, but the scale of the increase still surprised markets, sending shares lower in extended trading.

Alphabet’s outlook comes as investors have grown more wary of escalating AI costs across the technology industry. While demand for AI services continues to surge, concerns are mounting that returns may take longer to materialize, especially as competition intensifies and infrastructure requirements expand.

By comparison, Microsoft has indicated that its capital expenditures will ease sequentially after heavy recent spending, without offering a full-year figure. Meta has guided for 2026 capex of $115 billion to $135 billion, while analysts expect Amazon to continue increasing investment beyond its already sizable outlays. Alphabet’s forecast now sits at the top of that range, effectively setting a new ceiling for AI infrastructure spending among Big Tech.

Executives framed the spending surge as a necessary response to demand. Alphabet’s cloud unit, which houses most of its AI products, recorded a sharp acceleration in growth, with cloud revenue up nearly 48% year over year. Backlog also expanded rapidly, more than doubling from a year earlier to reach $240 billion by the end of the fourth quarter, reflecting long-term customer commitments for computing capacity.

Chief financial officer Anat Ashkenazi said the bulk of future investment will go toward expanding AI compute for Google DeepMind and meeting sustained cloud demand, alongside targeted investments to enhance consumer services and advertising returns. She noted that in 2025, roughly 60% of capital expenditures were directed toward servers, with the remainder spent on data centers and networking equipment — a mix that is likely to persist.

Management also emphasized progress on the product side. Alphabet said its Gemini AI app has grown to 750 million monthly active users, up significantly from the prior quarter. Chief executive Sundar Pichai highlighted the company’s partnership with Apple, under which Gemini models will play a central role in upgrading Siri, reinforcing Google’s position in consumer AI.

Despite those gains, Pichai acknowledged that infrastructure constraints remain the biggest challenge ahead. He pointed to limits around power, land and supply chains as the company races to scale capacity fast enough to keep up with demand. Alphabet has already moved to secure more assets, including a recent agreement to acquire data center operator Intersect.

The scale of Alphabet’s planned investment underscores a broader reality of the AI boom: infrastructure has become the most capital-intensive battleground. While Wall Street is increasingly cautious about the risks of unchecked spending, technology leaders appear convinced that falling behind on compute capacity would be even costlier.