Alphabet's AI Bet Pays Off: Google Cloud Surges 63% but Compute Can't Keep Up With Demand

For quarters, skeptics have asked whether Alphabet’s enormous AI infrastructure bet would ever produce numbers that justified the spending. On Wednesday, April 29, Alphabet answered — and the answer was unambiguous.

The company reported Q1 2026 revenue of $109.9 billion, a 22% year-over-year increase from $90.2 billion and well above analyst expectations. But the headline revenue figure was almost beside the point. The number that made investors send Alphabet shares up 6% in after-hours trading was the one buried in the cloud segment: Google Cloud grew 63% year-over-year to $20.02 billion, far exceeding the $18.05 billion analysts had penciled in and marking what CEO Sundar Pichai called “the strongest quarter ever for Google’s enterprise AI solutions.”

Net income came in at $62.58 billion, or $5.11 per share — an 81% surge from $34.54 billion in the year-ago quarter. Operating income reached $39.69 billion. By almost any measure, it was the quarter Alphabet’s AI narrative needed.

The Compute Constraint Problem

What made Pichai’s earnings call commentary particularly striking was not the growth — it was the admission of its limits. “We are compute constrained in the near term,” Pichai told analysts. “Our cloud revenue would have been higher if we were able to meet the demand.”

That sentence carries unusual weight. When a cloud provider says it cannot satisfy incoming orders, it is not explaining a weakness — it is describing a backlog problem born of strength. Google Cloud’s unearned revenue backlog nearly doubled quarter-over-quarter to over $460 billion, suggesting that enterprise customers are committing to Google’s AI infrastructure at a pace that even Alphabet’s aggressive data center buildout cannot immediately absorb.

The dynamic is structurally significant. It means the reported 63% growth rate may actually understate the underlying demand signal. Customers who want to run Gemini workloads on Google Cloud, access Google’s TPU capacity through the API, or use Vertex AI for enterprise deployments are being told they may have to wait. The $460 billion backlog is a reservation list for compute that has not yet been built.

This is why Alphabet raised its 2026 capital expenditure guidance on the call, moving from a prior range of $175 billion to $185 billion up to $180 billion to $190 billion. The additional $5 billion at the top end is the company’s signal that it is trying to build faster — even as Pichai acknowledged the gap between demand and available supply will persist “in the near term.”

Gemini Becomes a Revenue Engine

The AI monetization story at Alphabet is primarily a Cloud story, but the consumer layer is now pulling its weight in ways that were not visible a year ago. Gemini Enterprise saw 40% quarter-over-quarter growth in paid monthly active users in Q1 — a metric Alphabet had not previously disclosed with that level of specificity, suggesting the company wanted the market to see that enterprise adoption is compounding.

On the consumer side, the Gemini app has become a meaningful competitor in the conversational AI space, driven by the capabilities of Gemini 3.1 and the deep integration Alphabet has built across Search, Android, Chrome, and Google Workspace. The number of users activating AI features in Search grew significantly in the quarter, and Pichai tied this directly to Search revenue: “AI Overviews are genuinely increasing user satisfaction and query volume,” he told analysts.

Search and other revenues rose 19% to $60.4 billion, suggesting that the feared disruption to search monetization from AI-generated summaries has not materialized — and may actually be accelerating engagement. YouTube advertising came in at $9.9 billion, up 11%, showing steady if unspectacular growth as the platform navigates competition from TikTok and Meta’s Reels.

AI Infrastructure as the Primary Cloud Driver

“Our enterprise AI solutions have become our primary growth driver for cloud for the first time in Q1,” Pichai told analysts on the earnings call — a statement that marked a genuine inflection point in Alphabet’s cloud story.

For years, Google Cloud’s growth narrative was about chasing AWS and Azure in foundational cloud services: compute, storage, databases, networking. That race continues, and Google Cloud’s 63% growth (versus AWS’s 28% and Azure’s 40%) suggests Alphabet is gaining share. But the competitive moat Pichai was describing is something different: a situation in which Google’s AI capabilities — specifically its TPU architecture, the breadth of the Gemini model family, and deep vertical integrations like AlphaEarth and AlphaFold — are driving enterprise deals that competitors cannot easily replicate.

The backlog doubling to $460 billion is the clearest evidence that this narrative is being validated at the contract-signing level, not just the marketing level. Enterprise procurement cycles are long and painful; a company signs a multi-year Google Cloud AI commitment because it has done the evaluation and concluded that Google’s offering is competitively superior, not because it read a press release.

What Analysts Are Watching Next

The 63% Cloud growth rate will face a difficult comparable in Q2 2026, and analysts will be watching for any signs that the acceleration is decelerating. The compute constraint problem cuts both ways: if Alphabet cannot build data center capacity fast enough, the $460 billion backlog will take longer to convert to revenue, and management will need to explain how the capex increase translates to available capacity on a specific timeline.

The other key variable is AI pricing pressure. Google Cloud’s Gemini API pricing has been aggressive — the company has positioned Gemini Flash and Flash-Lite as the high-value, low-cost options in the enterprise AI market, aiming to gain volume at the expense of near-term margin. As long as the volume is there (and the $460 billion backlog suggests it is), the strategy is coherent. But margins will be scrutinized as compute costs rise.

Alphabet’s stock re-rating depends on whether the market concludes that the 63% Cloud growth and the compute-constrained backlog represent a durable structural advantage or a temporary surge that will normalize as competitors catch up. After Wednesday’s results, the bulls have considerably more data to work with.

The numbers, ultimately, speak plainly: in a quarter where the entire hyperscaler sector was on trial for its AI spending, Alphabet posted net income growth of 81%, the largest cloud beat in years, and a waiting list for AI compute that it literally cannot build fast enough to satisfy. Whether or not the compute constraint resolves cleanly, that is a problem most companies would envy.