AWS reports fastest growth in years as AI workloads surge, with data center capacity slated to remain constrained through 2028

Amazon Web Services is telling customers and investors that artificial intelligence demand is translating into record momentum for its cloud business, with growth at its fastest pace in nearly five years. In reporting carried by Yahoo Finance, the company linked the acceleration to a broad surge in AI workloads, ranging from frontier research efforts to more established enterprise deployments.

The same report also points to a supply-side bottleneck. AWS capacity, according to the account, is not expected to loosen quickly, with some capacity “selling out through 2028.” The comment underscores a familiar theme in the current cloud market: the limiting factor is often not software subscriptions but physical infrastructure, including data center space and specialized chips used to run AI training and inference.

While the report frames the demand as “massive,” it does not provide detailed revenue figures or quarter-by-quarter metrics in the materials available for review here. The thrust is that AWS is absorbing more computing demand than it can readily allocate, and that constraint is shaping customer planning timelines well into the late 2020s.

The AI workload mix described in the reporting is also notable. The acceleration is characterized as coming from both “frontier research labs” and “traditional” users, suggesting that AI spending is not limited to a narrow set of experimental teams. Instead, the implication is that organizations are building or scaling production systems that rely on the cloud for ongoing model development and for running inference workloads at scale.

Industry context matters because AI has changed how customers consume cloud services. Instead of using primarily general-purpose computing for batch workloads, many customers are now purchasing capacity oriented around machine learning pipelines. That includes GPUs and other accelerators, high-throughput networking, and storage systems that can handle rapid movement of large datasets. When those components are scarce, cloud providers can face longer fulfillment timelines even when demand is visible and recurring.

Amazon has previously emphasized that it is investing heavily in AWS infrastructure and AI capabilities, but what is most immediate in this report is The announcement on timing. The statement that capacity sells out through 2028 suggests AWS is prioritizing allocation decisions and capacity expansion, while customers may need to align new AI initiatives to procurement lead times.

Still, important specifics are not disclosed in the available reporting package. The materials here do not clarify whether the sold-out timeframe applies to all AWS regions, to particular services, or to specific capacity classes such as AI-focused GPU availability. The report also does not break out how much of the growth is attributable to training versus inference, or how customer spend is trending versus prior expectations.