"If you look at the valuations for all these AI names, they just can't all be accurate. You have memory makers at 3-5X …
Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-05-28
Atreides Management CIO Gavin Baker, overseeing $11B AUM, argues that current valuations across AI hardware names including memory makers, NVIDIA, custom chips, and optical networking cannot all be simultaneously correct.
Appears in
Extraction
Topics: ai-investmentsemiconductor-valuationsnvidiaai-hardware
Claims
- Current valuations across the AI hardware sector are internally inconsistent and cannot all be accurate at the same time.
- Memory makers are trading at 3-5X PE while NVIDIA is at a comparatively low PE, suggesting the relative pricing is distorted.
- The AI infrastructure investment landscape includes memory, custom chips, and optical networking as major components.
Key quotes
If you look at the valuations for all these AI names, they just can't all be accurate. You have memory makers at 3-5X PE. You have NVIDIA at a really low PE.