Amazon Cloud Growth Offsets Heavy AI Infrastructure Spending in Q2
Amazon reported strong second-quarter financial results on Thursday, driven by robust cloud computing performance that reassured investors despite escalating infrastructure spending. Net sales increased by 20%, while cloud revenue served as a major highlight, pushing Amazon’s stock up nearly 10% in after-hours trading.
What Happened
Despite investor preferences for spending restraint, Amazon continues to expand its data center investments significantly. For the fiscal year ended June 30, the company spent $173 billion on property and equipment—which includes GPUs, natural gas turbines, and land plots—up from $107.65 billion in the prior year. Additionally, Amazon raised its 2026 capital expenditure (capex) forecast from $200 billion to $220 billion.
To fund this expansion, Amazon tapped into its cash reserves, ending the quarter with $7.6 billion less cash than it held 12 months prior. This shift marked the company’s first period of negative free cash flow this year. However, market reaction remained positive due to strong operational performance at Amazon Web Services (AWS), where quarterly revenue grew 37% year-over-year to $42 billion, signaling strong market demand relative to infrastructure supply.
Key Highlights
- Amazon’s net sales grew 20%, driving stock gains of nearly 10% after hours.
- AWS generated $42 billion in quarterly revenue, representing a 37% year-over-year increase.
- Property and equipment spending reached $173 billion for the fiscal year ended June 30, compared to $107.65 billion the previous year.
- The 2026 capex forecast was raised from $200 billion to $220 billion.
- Cash reserves dropped by $7.6 billion over 12 months, resulting in negative free cash flow for the quarter.
- In-house chip developments include the Trainium TPU and Arm-based Graviton processor, aimed at improving cloud margins.
- Microsoft and Google also saw stock increases following strong cloud results, whereas Meta dropped 8% due to high spending without immediate clear revenues.
Why This Matters
Amazon’s strategy extends beyond construct physical data centers to designing proprietary silicon, such as the Trainium TPU and Graviton processor. While these chip projects do not appear directly in capex figures, they are intended to strengthen cloud operational margins over time.
During the Q2 earnings call, Amazon CEO Andy Jassy highlighted the business model, stating, “We see the AI business following very much the same margin trajectory we saw in the core business before. AWS and Amazon Bedrock can have a wildly successful business without its own frontier model, and the reason is that there’s not going to be a single model to rule them all.”
The broader market demonstrates a distinct trend: investors currently view cloud providers as a secure segment of the AI industry while remaining cautious regarding unproven AI labs and startups. However, cloud host revenue relies directly on customer expenditure. For instance, funding spent by AI labs like Anthropic translates directly into host revenues. If client spending slows down, cloud host revenues face potential risk across the sector.
What to Watch Next
Market observers will monitor whether end-user demand for AI applications remains strong enough to sustain the long-term infrastructure investment, referencing the broader market question posed by David Cahn regarding total industry demand.
Frequently Asked Questions
How much did Amazon AWS generate in Q2?
AWS generated $42 billion in revenue during the quarter, reflecting a 37% increase compared to the same period last year.
Why did Amazon’s cash flow turn negative?
Amazon’s cash reserves decreased by $7.6 billion year-over-year due to heavy spending on property, equipment, land, GPUs, and natural gas turbines for its data center network.
How did other major technology companies fare?
Microsoft and Google experienced stock gains following solid cloud results, while Meta saw its stock decline by 8% due to high capex and cash flow pressures without a distinct revenue engine.
Source: TechCrunch
