Dive Brief:
- Amazon Web Services reached $37.6 billion in revenue during the first quarter of fiscal year 2026, up 28% year over year, executives said during an earnings call Wednesday.
- Amazon CEO Andy Jassy said AWS remains confident in the capital expenditure investments the company is making in chips, data centers and other infrastructure. The tech giant committed $200 billion in CapEx throughout 2026.
- AWS is expanding its share of AI workloads, including through its deal with OpenAI, Dave McCarthy, program VP of cloud and infrastructure services at market intelligence firm IDC, told CIO Dive in an email. “The OpenAI-AWS partnership is arguably the most consequential infrastructure deal in enterprise technology in years — and enterprises should be paying close attention to what it means for their AI strategy,” he said.
Dive Insight:
AWS is banking on agents delivering the most enterprise AI value and plans to continue investing in the compute infrastructure necessary to support higher demand workloads.
In its recent partnership expansion with OpenAI, AWS announced limited preview availability of Amazon Bedrock Managed Agents powered by OpenAI, which allows enterprise customers to build AI agents with OpenAI frontier models on AWS.
AWS first announced plans to create this platform — what the companies called the Stateful Runtime Environment for agents in Amazon Bedrock — with OpenAI in February.
The Stateful Runtime Environment will enable organizations to build generative AI applications and agents at production scale, Jassy said during the company’s earnings call Wednesday.
“We believe that modern agentic applications will be stateful, and this new technology will rapidly accelerate agentic AI adoption,” Jassy said.
Along with enabling enterprises to build and scale AI agents, Jassy referenced AWS’ investments in agents for coding, software migration and business operations via tools such as Transform and Quick.
“While 2025 was about experimenting with LLMs, 2026 is about deploying autonomous agents that can execute workflows,” McCarthy said. “AWS is positioning itself as the ‘mission control’ for these agents, offering a managed environment where they can operate across software tools with the governance and security posture enterprises already have in place.”
Microsoft Azure and Google Cloud also saw significant revenue growth in the first quarter, 40% and 63% respectively, signaling the cloud infrastructure market is moving into a new phase as hyperscalers compete for enterprise spend, McCarthy said. All three have made significant investment commitments to building out AI infrastructure.
For Amazon and Google, that includes building custom-designed chips, with AWS seeing nearly 40% growth from the previous quarter in its chips business. AWS has entered into multigigawatt agreements with large language model providers Anthropic and OpenAI, as well as other companies, representing $225 billion in revenue commitments for its Trainium chips, Jassy said.
AWS also announced an agreement last week with Meta to use its Graviton processing cores — units inside CPUs that enable CPUs to support agentic AI workloads.
“AWS is to lay out cash for land, power, buildings, chips, servers and networking gear in advance of when we can monetize it, typically six to 24 months before we start billing customers depending on the component,” Jassy said. “However, these CapEx investments fund assets with many-year, useful lives — 30-plus years for data centers, five to six years for chips, servers and networking gear.”
