DENVER, Colo. (247marketnews.com) -- Rackspace Technology (NASDAQ:RXT) is making a bigger bet than simply selling access to GPUs: it wants to become the company enterprises call when artificial intelligence has to actually work in the real world.
Its latest move, joining NVIDIA’s (NASDAQ:NVDA) Cloud Partner Program and announcing Blackwell-powered infrastructure for regulated and sovereign customers, puts Rackspace squarely into one of the most consequential battles in enterprise technology: who controls AI after the demo ends?
The answer Rackspace is pitching is increasingly clear. NVIDIA supplies the accelerated computing foundation. Palantir supplies the enterprise AI and data software. Rackspace supplies the governed infrastructure, engineering and ongoing operations.
CEO Gajen Kandiah put the strategy bluntly: “Enterprise AI is becoming production infrastructure. At that point, the question is no longer simply who provides the GPU. It is who is accountable for operating the complete system.”
That is an important distinction in markets where healthcare, financial services, energy companies and governments increasingly need AI without surrendering control of sensitive data, models or deployment environments.
The NVIDIA relationship arrives after Rackspace has spent 2026 assembling the pieces of this strategy. In February, the company announced a strategic partnership with Palantir Technologies (NASDAQ:PLTR) to deploy Foundry and AIP in production under a governed operating model. In July, Rackspace and Palantir expanded that relationship, naming Rackspace a preferred partner for regulated and sovereign environments.
Now NVIDIA Blackwell becomes another major component of the stack.
Rackspace says its Institutional Sovereign Pod will combine NVIDIA's Blackwell architecture with Palantir Foundry and AIP, while Rackspace remains the operator. The pitch is less “rent us some GPUs” and more “give us the AI workload and let us own the complexity.”
That positioning could matter as enterprises move AI from experimentation toward production. Rackspace's own second-quarter results showed private-cloud revenue rising 5% year over year to $263 million, while CEO Kandiah said regulated enterprises were moving AI “from experimentation into production.”
There is another revealing piece to the strategy: Rackspace isn't tying itself to one chipmaker. In June, the company announced a definitive agreement with Advanced Micro Devices (NASDAQ:AMD) for a phased deployment of an initial 30 megawatts of AMD-based compute across its data centers beginning in late 2026 through 2028.
That makes the NVIDIA announcement look less like a one-off partnership and more like a deliberate infrastructure land grab.
Rackspace is trying to sit above the NVIDIA versus AMD hardware fight and sell the thing enterprises may ultimately value most: governed, managed AI that someone is actually accountable for keeping online.
Important Editorial Note: 247 highlights companies approaching significant catalysts and inflection points. This report reflects information available at the time of publication. Since developments can occur rapidly, readers should independently verify current information and review all company filings and disclosures.