Agentic AI Drives CPU Demand, Lifting Intel and AMD Shares
Meta's Muse agent boosts CPU and memory demand, triggering double-digit gains in chipmakers and optical networking stocks.
Key points
- Meta's Muse agent drives demand for CPUs and memory, boosting Intel, AMD, and Arm shares.
- Nebius raises CPU-only cloud instance prices by 25% due to tight server CPU supply.
- NVIDIA targets $20 billion in CPU revenue, while memory and optical stocks rally on cluster demand.
The AI hardware sector rebounded sharply as demand for agentic AI expanded compute requirements beyond GPUs. Intel, Advanced Micro Devices, and Arm Holdings posted double-digit intraday gains, driven by a new revenue stream from cloud-based virtual environments that rely heavily on central processing units and memory. This shift reverses the recent selloff caused by fears of an AI market slowdown, with macro tailwinds from falling oil prices and a retreat in 10-year Treasury yields further easing pressure on high-valuation tech stocks.
Meta Platforms’ Muse agent, which reached the top of the U.S. iPhone free-app chart ten days after launch, exemplifies this demand shift. Each agent operates in its own cloud-based virtual environment, utilizing CPUs, memory, and storage for task orchestration and code execution, while GPUs handle model inference. This operational model expands the addressable market for traditional cloud infrastructure, creating a new demand story that benefits the broader compute ecosystem rather than just AI accelerators.
Server CPU Supply Tightens
Manufacturers are already responding to the increased load on server CPUs. Prices for Intel and AMD server processors have risen significantly this year, accompanied by extended lead times that signal tightening supply. Nebius, a cloud infrastructure provider, plans to raise prices for certain CPU-only cloud instances by 25% starting October 1. This pricing action reflects the growing scarcity of compute resources capable of handling the parallel workloads required by agentic AI systems.
Broad Compute Ecosystem Benefits
The upside extends across the chip supply chain. Arm Holdings stands to gain as hyperscalers like Amazon, Alphabet, and Microsoft adopt Arm-based architectures for in-house CPUs such as Graviton, Axion, and Cobalt to handle agent workloads. NVIDIA is also expanding into this segment with its next-generation Vera CPU, an Arm-based architecture for which the company has identified a nearly $20 billion standalone revenue opportunity this fiscal year. This diversification indicates that CPU revenue is becoming a critical component of the AI business model for major semiconductor firms.
Memory And Optical Stocks Rally
Strength in the AI infrastructure stack is visible in memory and optical networking sectors. Micron Technology and SanDisk rallied as demand for High Bandwidth Memory, DRAM, and enterprise SSDs grows alongside AI server deployments. Simultaneously, optical networking firms including Lumentum, Coherent, Ciena, Applied Optoelectronics, and Corning moved higher. Investors are pricing in rising demand for high-speed connectivity, as larger AI clusters and increased inference workloads require greater network bandwidth and memory consumption. According to reporting from moomoo.com, this broadening rally suggests that the AI trade is no longer confined to GPUs but is spreading across the entire infrastructure stack.
