NVIDIA CEO Jensen Huang and Dell CEO Michael Dell took the Dell Technologies World keynote stage on May 18, 2026 to unveil the latest updates to the Dell AI Factory with NVIDIA — a full-stack platform for autonomous agents, from deskside workstations to data center racks.
The Keynote
“We’ve now arrived at the era of useful AI, which is the reason why demand is going parabolic, utterly parabolic,” Huang said. “What took months now takes weeks, and what took weeks now takes days.”
Michael Dell sized the stakes: Worldwide AI infrastructure spending could reach $3-4 trillion by 2030, with token consumption projected to grow 3,400 percent in the same window. He welcomed Huang on stage with the NVIDIA portfolio behind him — from a deskside Dell Pro Max with GB10 workstation to a Dell PowerRack with NVIDIA Vera Rubin NVL72.
A New AI Factory for the Agentic Era
The accelerated computing news leads the refresh. The Dell PowerEdge XE9812, built on NVIDIA Vera Rubin NVL72, delivers up to 10x lower cost-per-token than NVIDIA Blackwell for massive-scale agentic AI inference. It is joined by PowerEdge XE9880L, XE9885L and XE9882L servers — the first Dell systems on NVIDIA HGX Rubin NVL8, supporting up to 144 GPUs per rack with 100% direct liquid-cooled compute.
The CPU story is equally central: Dell PowerEdge M9822 and R9822 servers bring NVIDIA Vera CPUs to the AI factory. With 1.2 TB/s memory bandwidth, Vera completes agentic workloads 50% faster than x86 processors, and data engine Starburst delivers 3x faster query throughput on Vera CPU for large-scale SQL analytics.
Agents and Models, Securely
Dell’s AI adoption survey found that 67% of AI workloads now run outside the cloud — on premises, on device, at the edge or in colocation. The answer relies on NVIDIA Confidential Computing, delivered with Fortanix, Google, Red Hat and others, as the foundation for deploying frontier models inside the enterprise without exposing model IP or data.
- Google Distributed Cloud with Gemini 3.0 is available in preview on Dell PowerEdge XE9780, accelerated by NVIDIA Blackwell and secured by Confidential Computing
- SpaceXAI models are coming on premises with model weights and data protected end to end
- NVIDIA Nemotron open models run natively for enterprises that want open-weight models tuned to their domain
Ecosystem and Software Partners
OpenAI Codex will connect with the Dell AI Data Platform to bring agents closer to internal codebases and workflows. Palantir’s sovereign AI OS reference architecture now runs on Dell infrastructure, and NVIDIA OpenShell is supported across the entire Dell AI Factory. Customers featured on stage include Lilly, Samsung, Honeywell and Hudson River Trading.
What This Means
The message from Monday’s keynote: enterprise AI has moved past pilots into agentic inference at scale. With Vera Rubin-based systems, confidential computing and a growing open-model library, Dell and NVIDIA are positioning the on-premises AI factory as the platform where the next decade of enterprise agents will run.