Back to News
Technology
Jun 8, 202615 views2 min read

Intel Unveils Rackscale AI Infrastructure at Computex 2026 to Support Agentic Workloads

Intel announced new rackscale AI infrastructure at Computex 2026, built in partnership with SambaNova and Foxconn. The systems use Intel Xeon 6+ processors on the 18A process node and SambaNova RDUs to optimize performance for agentic AI workloads. The announcement signals a shift in enterprise computing as AI agents move from experimental pilots to production deployments.

Intel Unveils Rackscale AI Infrastructure at Computex 2026 to Support Agentic Workloads

Intel announced new rackscale AI infrastructure at Computex 2026, developed in partnership with SambaNova and Foxconn. The systems are designed specifically for agentic AI workloads, where AI systems reason, plan, and execute multi-step tasks without constant human oversight.

The hardware uses Intel Xeon 6+ processors built on the company's 18A process node, paired with SambaNova Reconfigurable Dataflow Units to optimize performance density. Intel says the combination delivers better efficiency for the types of workloads that agentic AI systems require compared to traditional GPU-based setups.

The announcement reflects a broader shift in enterprise computing. Agentic AI has moved from experimental pilots to production deployments at a growing number of companies. Microsoft, Amazon, and other major cloud providers are integrating agentic capabilities into their platforms.

Amazon's DeepFleet AI system, for example, now coordinates over a million robots across its warehouse network to optimize logistics efficiency. Uber is testing robotaxi programs in Munich using Nvidia's Drive Hyperion platform.

Intel's move into rackscale AI infrastructure is also a response to competitive pressure from Nvidia, which has dominated the AI hardware market. By targeting the orchestration layer of agentic workloads, Intel is positioning its CPUs as essential components rather than secondary processors.

Analysts at Deloitte note that as agentic workloads proliferate, the ratio of CPUs to GPUs in data centers is shifting back toward a more balanced split, returning the CPU to a more prominent role in AI infrastructure.