Key Takeaways
- This year’s AWS Summit, held in NYC, focused on how the internet is transforming to accommodate the next-gen era of agentic AI.
- Major announcements included AgentCore, vector storage, and a dedicated AI-specific marketplace.
- AWS VP Swami Sivasubramanian emphasized that AI agents will still depend on broader infrastructures — so opportunities abound for providers to join the AI ecosystem.
This year’s AWS Summit in New York City was all about one thing: The future of the internet is going from cloud-native to AI-native.
During the 90-minute keynote, Swami Sivasubramanian, AWS’s VP of Agentic AI, described this new era of AI-powered software as “the most impactful change we’ve seen since the dawn of the internet.”
“It upends the way software is built,” said Sivasubramanian. “It also introduces a host of new challenges to deploying and operating it, and potentially most impactfully, it changes how software interacts with the world — and how we interact with software.”
Surely, others will follow suit. And in the meantime, it’s time for providers across the board (be it hosts, network, or infra) to get ready to host these upcoming agentic workloads.
What Matters Inside AWS’s AI Toolkit
Instead of highlighting the usual cloud topics like storage and scalability, AWS used the Summit to share what it’s doing to shift AWS into the agentic AI era.
“You can really see the innovation coming out of our engineering teams to help enterprises not only develop these agents, but also make sure that they’re deploying them in a secure, reliable way,” said Ben Schreiner, Head of AI and Modern Data Strategy at AWS, in an interview with theCUBE.
AgentCore One is a key development to watch — a modular suite designed to deploy AI agents both securely and at scale. It’ll be a groundbreaking tool for developers, and hosts that support developer environments should ensure their infra is ready to support it.
Other significant updates include:
- A new agentic AI category in AWS Marketplace, where devs can browse, buy, and deploy AI agent tools from companies like Anthropic and IBM. Hosts running on AWS can act as the middle layer, or better yet, integrate these tools into managed, ready-to-deploy packages.
- AWS’s Nova foundation models are now tunable in SageMaker, meaning developers can customize models for specific industries. It’s a good opportunity for hosts that work with certain verticals, like healthcare or finance.
- TwelveLabs, a video AI company, is now available in Bedrock, offering features like scene detection, summarization, and auto-captioning. Instead of relying on players or embeds, this integration makes video content more dynamic and accessible.
AWS also announced an additional $100 million investment toward its Generative AI Innovation Center and startup incentives.
“This all comes together to make AWS the best place to build and deploy agents,” Sivasubramanian said. “We are still in the early days of agents, but with these innovations, we are charting the path to enable you to build a bright future.”
It’s a Web Hosting Wake-Up Call
In his keynote, Sivasubramanian emphasized that developing AI agents isn’t just about deploying your favorite LLMs; it also demands deep expertise within the existing infrastructure.
“No organization can be an expert in everything — nor should they be,” he said. “Just as today’s software ecosystem thrives on third-party APIs, tomorrow’s AI agents will need to integrate specialized capabilities from across organizations, providers and systems.”
In other words, like most software, it won’t operate in a vacuum.
Fortunately, many providers — Cloudflare, Akamai, Fastly, Netlify, DigitalOcean, and others — are already implementing AI-nativity so hosts don’t have to build everything from scratch.
Still, it’s fascinating to look at how far web development has come in the past few decades: From manual-coding in the ‘90s to the rise of cloud infrastructure in the 2010s, hosting has evolved from “Where do I upload my files?” to “How do I run smart apps without having to build from scratch?”
And according to AWS, the answer is clear: agentic AI-ready infrastructure.
It’s as AWS DevTools Hero Bhuvaneswari Subramani said: “We are moving into an AI-driven era where AI takes care of higher order activities.”
