Wall Street Calls a Changing of the Guard in AI Chips as AMD, Intel and Micron Sprint Past Nvidia

Wall Street has decided the next leg of the AI buildout will not be carried by Nvidia alone. In a note that quickly rippled through trading desks on May 8, 2026, Mizuho analyst Jordan Klein described the past week as a clear changing of the guard in AI semiconductors, with money rotating out of the dominant GPU maker and into a broader basket of chip and memory names that had spent most of the cycle in its shadow.

The numbers tell the story bluntly. Advanced Micro Devices and Intel each notched roughly 25 percent gains over the period, while Micron, the memory specialist suddenly central to every accelerator roadmap, surged more than 37 percent. Corning, the 175-year-old glass and fiber-optic supplier now embedded in Nvidias data center plumbing, tacked on about 18 percent. All four have more than doubled this year, with Intel up over 200 percent on the back of fresh foundry interest and government-backed manufacturing commitments.

Nvidia, by contrast, is up just 15 percent year to date, barely ahead of the Nasdaq, even after an 8 percent rally this week. That is a striking gap for a company that defined the first phase of the generative-AI investment cycle and one whose equity stakes across the industry crossed 40 billion dollars earlier this month. The shift is not a verdict against Nvidia so much as a recognition that the next bottleneck is no longer raw GPU supply.

Driving the rotation is a structural change in how AI workloads are being deployed. As applications move from chatbots to long-running agents and multimodal pipelines, demand is widening to high-bandwidth memory, advanced packaging, networking glass, and CPU host platforms. Micron, whose market capitalization broke 800 billion dollars for the first time this week after a more than 750 percent run over the past year, has become the clearest beneficiary of that supply-demand mismatch as a global memory shortage collides with frontier model training cycles.

For investors, the signal is that AI infrastructure is maturing into a more diversified stack. The hyperscaler capex flowing through Nvidia is now spilling into every adjacent layer, from AMDs MI400 generation to Intels Gaudi and 18A foundry slots, to Microns HBM4 ramp and Cornings optical interconnect business. Whether the rotation persists will depend on Nvidias next earnings print and on whether the new entrants can convert pipeline wins into sustained revenue, but for one week at least, the AI trade had a notably wider footprint.