Google’s Q2 2026 results prove AI spend is now the story

Google’s Q2 2026 earnings show that AI infrastructure spending now matters more than search.

Google’s Q2 2026 report says the market has stopped valuing search as the whole story and started pricing AI capacity as the main one.

Google’s growth engine is still search, but search is no longer the edge

Alphabet said search revenue reached $63.3 billion in the quarter, just shy of the $63.4 billion Wall Street expected. That is not a collapse, but it is a warning sign for a business that still carries the company’s cash generation and investor narrative. When a mature core line lands exactly in line or slightly light, the market stops rewarding stability and starts asking what comes next.

The real signal is that search grew 17% year over year, which is solid by any standard and still enormous in absolute dollars. Yet investors barely cared because the company itself pointed them toward a tougher third quarter, with finance chief Anat Ashkenazi saying Google will begin lapping a faster search comparison from last year. In other words, search is still strong, but it is no longer the part of Alphabet that changes the valuation debate.

Cloud is now the proof point investors cannot ignore

Google Cloud revenue jumped 82% to $24.8 billion, one of the clearest signs that Alphabet’s AI infrastructure bet is already producing commercial output. That kind of growth is not a vanity metric. It shows that customers are paying for Google’s compute, storage, and model access at a scale that justifies the company’s aggressive buildout.

The cloud number matters even more because management framed supply as the bottleneck, not demand. Ashkenazi said Google is still in a supply-constrained environment and will expand third-party capacity in Q3 as a bridge. That is the kind of comment investors should treat seriously: demand is strong enough that the constraint is physical capacity, and that is exactly why the market should be watching capex, not just revenue beats.

Capex is the new scoreboard for Alphabet

Alphabet raised its 2026 capex forecast to $195 billion to $205 billion, up from $180 billion to $190 billion. That is a massive jump, and it tells you the company is choosing speed over restraint in the AI arms race. The spending is not abstract either. Google said 60% of infrastructure investment went to servers and 40% to data centers and networking equipment.

Q2 capex alone hit $44.9 billion, up 100% year over year, and most of it went to technical infrastructure supporting AI. That is the clearest evidence that Alphabet is turning cash flow into compute at a pace Wall Street can no longer treat as optional. The stock reaction after hours makes sense: investors saw that the company is winning demand, but also that the bill for winning is rising faster than they wanted.

The counter-argument

There is a strong case that the market is overreacting to capex and underweighting the scale of the opportunity. Alphabet’s cloud growth is accelerating, its Gemini App has 950 million monthly active users, and its AI coding tool Antigravity has 2.4 million weekly active users. Those are real signs of product adoption, not just hype. If AI demand keeps compounding, heavy spending today can lock in a much larger platform later.

There is also a balance-sheet and portfolio argument. Alphabet still has a dominant ad business, a fast-growing cloud arm, and stakes in valuable private assets such as Anthropic and SpaceX. The company’s “other income” line added $99 billion, which reinforces the point that Alphabet has multiple ways to absorb a period of aggressive investment without breaking the model.

That argument is valid, but it does not change the present reality: investors are not being asked to fund a theoretical future, they are being asked to underwrite a very large near-term cash outlay in a supply-constrained market. The right view is not that capex is bad, but that Alphabet has made capex the primary variable in the stock. Until the company proves those dollars translate into durable margin expansion, the market will keep treating every spending hike as a risk factor.

What to do with this

If you are an engineer, PM, or founder watching Alphabet, read this quarter as a playbook for the AI era: distribution still matters, but capacity, cost, and product adoption now decide who wins. Build for usage that can scale into revenue, watch your infrastructure unit economics as closely as your growth charts, and assume investors will reward proof of constrained demand solved by disciplined execution, not just big model claims.