AWS AI Fuels Amazon's Q2 Surge

Amazon's cloud division, AWS, experienced a significant revenue acceleration in the second quarter, with growth hitting 36.7% year-over-year. This marks the fastest growth rate in 18 quarters, fueled by a symbiotic relationship between artificial intelligence services and foundational cloud infrastructure. Amazon CEO Andy Jassy noted that AWS is now a $169 billion annualized revenue run rate business, a scale that would place it 24th on the Fortune 500 if it stood alone. According to Amazon News, the company's custom chips business also surpassed a $25 billion annualized revenue run rate, growing at triple-digit percentages.

AI and Core Services Drive Growth

Jassy attributed AWS's appeal to its broad capabilities, its position as the preferred location for AI inference due to proximity to data and applications, and its robust security and operational performance. He detailed how AI growth bolsters core services, particularly through the increased demand for CPUs in post-training, reinforcement learning, and agent tool use. AWS's Graviton chips, offering superior price-performance, are a key advantage here. The need to store vast AI datasets and run vector databases also plays to AWS's strengths, with the company claiming a significant lead in these core infrastructure areas.

The company is seeing strong adoption across its AI stack. Jassy reiterated Amazon's stance that technically competent companies will build their own specialized foundation models, not necessarily frontier models. Amazon's SageMaker service simplifies this process, while Amazon Bedrock provides a high-performance, cost-effective inference service with leading models and essential governance. Bedrock is also expanding rapidly.

Agentic Services Take Center Stage

Building and deploying AI agents remains a complex challenge for many enterprises. Jassy highlighted Bedrock AgentCore, which offers managed infrastructure components like policies, payments, and web search to simplify agent development and deployment. Features like Strands are further accelerating the creation of these agents.

Beyond custom agent development, turnkey solutions are gaining traction. Coding agents, such as Amazon's Kiro, are proving cost-effective and seeing increased usage. Amazon Quick, an AI work companion, is also gaining momentum, managing digital workloads across various applications and SaaS tools with enterprise-grade security controls. In Q2, Quick introduced autonomous agents and expanded its integrations, with major companies like 3M and BMW already adopting it.

AWS also offers agentic services like Amazon Connect for call centers and AWS Transform for software migration. A new offering, AWS Continuum, addresses the growing concern of AI-assisted security vulnerability discovery. Continuum ingests existing vulnerability backlogs and uses frontier models to scan, prioritize, validate, and remediate code flaws using agents and business context, aiming to reduce false positives and recommend fixes.

The rapid expansion of AWS AI services, particularly in areas like agent development and security, positions Amazon as a key player for enterprises navigating the complexities of generative AI adoption. This focus on both foundational infrastructure and specialized AI tools is a critical differentiator.