Agentic AI Revives CPU Demand, Lifts Intel and AMD

The rise of agentic AI—autonomous systems that execute multi-step tasks—is reshaping data center hardware demand, pushing CPU and memory markets into a new growth cycle. Intel and AMD shares surged as their data center revenues hit record highs, signaling that AI infrastructure investment is broadening beyond GPU dominance.

Agentic AI Requires CPUs for Multi-Step Task Execution

Unlike traditional chatbots, agentic AI systems perform sequences of actions—searching databases, making API calls, and verifying permissions—which require the sequential processing power of CPUs to complement GPU-based inference. This shift is driving renewed demand for server CPUs even as GPU deployments continue to expand.

AMD Data Center Revenue Surges 57% YoY; Intel Up 22%

AMD’s data center segment posted a 57% year-over-year revenue increase, while Intel saw a 22% rise, both reaching all-time highs. The gains reflect not only GPU-adjacent workloads but also the growing need for CPU cycles in agent-driven workflows.

Memory Demand Expands from HBM to DDR5 and Server DRAM

The agentic AI trend extends the memory upcycle, with demand broadening from high-bandwidth memory (HBM) to DDR5 and server DRAM. Analysts also point to the potential emergence of CXL (Compute Express Link) infrastructure to handle the increased memory bandwidth and capacity requirements.

AI Hardware Investment Broadens Beyond GPUs

Agentic AI is acting as a new growth engine for the entire AI infrastructure stack. The dual demand for CPUs and memory indicates that hardware procurement is expanding beyond GPUs, creating opportunities for broader semiconductor and memory manufacturers.

• Agents need CPUs for sequential logic and orchestration
• AMD data center revenue up 57% YoY; Intel up 22%
• Memory demand expands from HBM to DDR5 and CXL