Wall Street Raised Amazon Targets After AWS Grew 37%. The Number to Watch Is $220 Billion.

Why Amazon's quarter changed the setup

This quarter changed the argument. Amazon did not just post a clean beat. AWS grew 37%, AWS revenue reached $42.2 billion versus $40.54 billion expected, and the rally was big enough to add more than $340 billion in market value. At least 15 brokerages also raised their price targets. That combination is what turns a good quarter into a new base case.

Why this week mattered

Earlier this month, investors were still asking whether AI spending was being front-loaded with too little proof. Reuters described it as a "make-or-break" week for big-tech confidence. Amazon, Microsoft, and Alphabet all reported strong cloud growth in the same stretch. For Amazon, the key tell was AWS. A 37% growth rate is hard to dismiss as a one-quarter anomaly.

The number now under the microscope

From here, the debate is simpler: $220 billion. Amazon raised its 2026 capital expenditure outlook to that level from $200 billion. Bulls now have proof of demand; bears will argue the spending ticket just got bigger.

That is why the capital-expenditure figure matters most now. Management says the extra spending is serving existing demand, and some analysts pointed to capacity already reserved by customers. The next test is whether AWS keeps converting that demand into revenue fast enough to justify the next leg of spending.

Why AWS accelerated: AI demand is lifting core workloads

The important shift is not just that AWS accelerated, but why it accelerated.

AI demand is pulling more of the cloud stack with it

AWS is now at a $169 billion annualized revenue run rate. Its Chips business run rate over $25 billion and AI revenue run rate, also over $25 billion, are both growing at triple-digit year-over-year rates. That combination changes the story. This looks broader than a short-lived model-training spike.

Management's point is straightforward: AI demand is pulling more core cloud usage with it. Post-training work, reinforcement learning, and agent tooling often rely heavily on CPUs, not just AI accelerators. That matters for Amazon because its Graviton chips offer up to 30-40% better price-performance than other options. In practical terms, once companies train a model, they still need compute, memory, databases, and tooling to put it to work.

Why wallet share can keep widening

Customers also want inference close to their existing applications and data, and more of that workload already lives in AWS. Management also pointed to what it described as the strongest security and operational performance. The business logic is simple: the more data and workflows already sit in AWS, the less reason customers have to fragment spending elsewhere.

Why the mix looks more durable than a passing trend

Management also said many companies are likely to build their own smaller, proprietary models rather than rely only on frontier models. That points to a longer tail of demand across storage, databases, inference, and surrounding services.

The key watchpoint is still adoption. If broad AI adoption slows or customers remain stuck in pilots, the premium narrative would need to cool. For now, though, AWS is growing in both AI and core workloads, which makes the spending case look more like infrastructure buildout tied to real demand.

The bull case, the main risk, and what matters next

The rerating has already done some of the hard work. After the report, the stock soared approximately 15.2% by Friday's close, and it now sits in the middle of its 52-week range and near its 200-day moving average. The easy "nobody believed it" phase is gone. Bulls can still argue the setup improved, but bears now have a cleaner test: when a leader rallies this hard, future returns depend more on follow-through.

Where the bull case still holds

The strongest bull point is that Amazon says much of 2027 capacity and some 2028 capacity already reserved by customers. That shifts the question from whether demand exists to whether Amazon can build and fill that capacity quickly enough to keep converting demand into revenue.

There is a second layer to the case. AWS is not leaning on one showpiece AI product. Its Chips business run rate over $25 billion and AI revenue run rate, also over $25 billion, both point to customers buying more of the stack, not just a short burst of GPU time.

The catch: the cash-flow bill is arriving now

The main pressure point is cash flow. Amazon's trailing 12-month free cash flow was negative $7.6 billion, compared with positive $18.2 billion a year earlier. That is a sharp reversal. Investors can tolerate heavy spending if it is buying something valuable, but they usually do not tolerate a wide cash-flow gap for long.

What to watch next quarter

The next report needs to prove a short list of things:

• AWS growth remains elevated, rather than reverting after one strong quarter.
• AI demand continues feeding into core workloads, especially CPU-heavy tasks.
• Reserved capacity starts supporting cash generation instead of widening the cash-flow gap.
• Management can show better returns on the extra spending.

Demand now looks credible. The next test is whether revenue and cash flow catch up quickly enough to justify the next move higher.