AMD and Rackspace sign deal to deploy 30 MW of AI compute for regulated industries

AMD and Rackspace Technology signed a definitive agreement on June 16, 2026 to deploy an initial 30 MW of AMD AI compute across Rackspace's global data centers, AMD said in its newsroom. The phased rollout runs from late 2026 through 2028 and operationalizes a memorandum of understanding the two announced on May 7, 2026.

The footprint will pair AMD Instinct GPUs — including the MI355X and MI350P and future successors — with AMD EPYC CPUs inside an integrated Enterprise AI Cloud architecture. Rackspace plans four capabilities on top of it: Enterprise AI Cloud, an Enterprise Inference Engine, Inference as a Service, and Bare Metal AMD Instinct.

The partnership targets regulated industries, with healthcare providers expressing early interest in clinical AI and inference at scale, AMD said. Rackspace Chief Executive Gajen Kandiah said the deal delivers "a governed AI stack with one accountable partner from silicon to outcomes." Dan McNamara, AMD senior vice president and general manager of compute and enterprise AI, said it would help regulated enterprises deploy high-performance AI infrastructure with the openness, scalability and accountability needed to run AI at enterprise scale.

Separately on June 16, Rackspace disclosed a 15% global workforce reduction in an SEC filing, which it expects to generate $75 million to $85 million in annualized run-rate savings against one-time restructuring costs of $14 million to $19 million. Most affected employees were notified around June 10, with further reductions expected over the following six months. Rackspace shares rose roughly 16% on the day.

The companies cautioned that individual deployment authorizations and certain commercial terms remain subject to further agreement, and that third-party financing depends on availability. Deployment volumes beyond the initial 30 MW were not specified.