How AWS and NVIDIA Are Powering Agentic and Physical AI
The AWS NVIDIA partnership just entered a major new phase, as the two companies announced a significant expansion of their long-standing collaboration to meet surging global demand for AI infrastructure.
As the broader AI buildout continues accelerating, compute power has emerged as one of the fastest-growing resources in the world. Consulting firm McKinsey estimates that data centers globally will require roughly $6.7 trillion by 2030 just to keep pace with unprecedented demand for computing capacity.
A Massive Expansion in GPU Deployment
Building on already rapid customer adoption of NVIDIA-accelerated compute through AWS, the two companies now plan to deploy two million additional NVIDIA GPUs across AWS’s global infrastructure. “NVIDIA and AWS have built one of the great growth engines of the AI era, and demand is running ahead of every forecast,” said Jensen Huang, founder and CEO of NVIDIA.
This latest announcement builds directly on an earlier commitment from March, when AWS first revealed plans to add more than one million NVIDIA GPUs starting in 2026. Since then, demand has reportedly exceeded even those ambitious projections, prompting AWS to commit to deploying an additional two million NVIDIA Blackwell Ultra, Rubin, and Rubin Ultra GPUs across its global infrastructure through 2027 and 2028, including dedicated AI factories.
Powering Physical and Agentic AI
The AWS NVIDIA partnership now spans a considerably broader range of technology areas, including AI factories, CPUs, networking, open models, data processing, and robotics. NVIDIA has framed the collaboration as delivering co-engineered AI solutions designed to help customers accelerate AI development and deployment at what the company describes as unprecedented scale.
Matt Garman, CEO of AWS, explained the strategic reasoning behind the deepened collaboration. “Customers want the freedom to choose the best tools for their AI workloads, and they want confidence that everything works seamlessly together,” Garman said. “That’s why we’ve invested deeply with NVIDIA to make AWS the best place to run NVIDIA AI technologies, optimising performance across our infrastructure from networking and security to deployment.”
Huang echoed that sentiment, framing the announcement as a natural evolution of a partnership that now spans 16 years. “For 16 years, we have scaled NVIDIA computing in the cloud together,” he said. “Now, we are expanding our partnership across the full stack, GPUs, CPUs, networking, open models and software, to make agentic and physical AI real at an unprecedented pace and scale that only AWS and NVIDIA can deliver.”
What the Expanded Collaboration Actually Includes
The AWS NVIDIA partnership outlines several specific components in this latest expansion. This includes extending NVIDIA NVLink Fusion technology with custom NVIDIA high-bandwidth memory, and building dedicated AI factories for the US government.
The collaboration also involves integrating NVIDIA’s platform directly with the AWS Nitro System and Elastic Fabric Adapter to strengthen both security and reliability. Additionally, the companies will continue supporting NVIDIA Nemotron open models across Amazon Bedrock and Amazon SageMaker, while accelerating data processing and vector indexing capabilities on Amazon EMR and Amazon OpenSearch through NVIDIA’s cuDF and cuVS CUDA-X libraries.
Robotics represents another key focus area, with Amazon Robotics adopting NVIDIA’s broader physical AI platform, including Jetson, Omniverse, and Isaac technologies, to further advance automated robotics workloads across Amazon’s operations.
Industry Reaction to the Announcement
The expanded partnership has already drawn attention from companies building directly on AWS infrastructure. Skylar Graika, CTO and co-founder of AI-powered body camera company Patrol 6, which operates on AWS, welcomed the development publicly on LinkedIn. “It’s exciting to see AWS and NVIDIA making AI production-ready at scale,” Graika wrote. “Moving from pilot to full deployment is where so many teams get stuck, and these integrations could really accelerate real-world impact.”
That sentiment reflects a broader industry challenge the expanded stack appears specifically designed to address: helping companies bridge the persistent gap between early AI pilot programs and genuine full-scale production deployment.
NVIDIA’s Broader Partner Ecosystem
While the AWS NVIDIA partnership remains central to both companies’ AI strategy, NVIDIA maintains partnerships with several other major hardware and cloud providers. Taiwan Semiconductor Manufacturing Company, the world’s leading pure-play semiconductor foundry, manufactures NVIDIA’s most advanced AI and graphics chips, including its Blackwell and Rubin architectures, using cutting-edge fabrication techniques and advanced chip packaging technology.
Dell Technologies collaborates with NVIDIA on integrated enterprise solutions like the Dell AI Factory with NVIDIA, combining NVIDIA GPUs, networking, and AI software with Dell’s server infrastructure to offer scalable AI deployment options for enterprises running on-premises or hybrid cloud environments.
Foxconn, the world’s largest electronics contract manufacturer, works with NVIDIA to produce high-density AI server systems, while also integrating NVIDIA’s physical AI ecosystem, combining Omniverse digital twins and Isaac robotics, to build next-generation automated factories and smart manufacturing facilities.
As the AWS NVIDIA partnership continues expanding across this broad technology stack, the two companies appear positioned to remain central players in how enterprises, governments, and research institutions scale AI infrastructure over the coming years.
Source: This article is based on reporting by Adam Pond for FDi Intelligence.