Rackspace and AMD commit to 30MW AI cloud for regulated sectors

Rackspace Technology and AMD have formalized a partnership to deploy 30 megawatts of AI cloud infrastructure, specifically designed for enterprises operating under strict regulatory oversight. The phased rollout, scheduled to begin in late 2026 and continue through 2028, will leverage AMD’s Instinct MI355X and MI350X GPUs alongside EPYC CPUs, integrated into Rackspace’s Enterprise AI Cloud platform. This architecture aims to provide a single operational model for AI workloads, eliminating the need for customers to manage disparate infrastructure layers from multiple vendors independently.

The collaboration targets sectors where compliance, data sovereignty, and operational transparency are non-negotiable, such as healthcare and financial services. Rackspace’s pitch centers on consolidating accountability under one cloud operator, a critical factor for enterprises navigating regulatory scrutiny over data handling and system governance. The agreement also solidifies four service components initially outlined in a prior memorandum: Enterprise AI Cloud, Enterprise Inference Engine, Inference as a Service, and Bare Metal AMD Instinct. These offerings span from physical compute to fully managed inference services, allowing customers to engage at the cloud layer that best suits their operational needs.

Strategic alignment for regulated industries

The partnership extends beyond a hardware supply arrangement, with both companies committing dedicated sales and marketing teams to jointly pursue customers in regulated industries. This co-selling approach reflects a shared accountability model, where Rackspace and AMD will collaborate to ensure the cloud infrastructure meets compliance and performance requirements. For AMD, the deal strengthens its position in the enterprise AI cloud market by tying its processors to a managed cloud environment, rather than relying solely on chip sales. This shift addresses a key customer concern: whether the entire AI cloud stack can meet regulatory standards, rather than focusing on individual hardware components.

Capacity and market implications

At full deployment, the 30-megawatt AI cloud footprint will provide substantial capacity for regulated enterprises transitioning AI workloads from pilot phases to production environments. The infrastructure is positioned to support large-scale inference and clinical AI applications, where data governance and operational transparency are critical. The partnership’s focus on co-selling and shared accountability could set a precedent for future collaborations in the AI cloud space, particularly in industries where compliance is a primary consideration.