Nvidia stock valuation falls to pre-AI boom levels in 2026 - qz.com

NVIDIA's stock has fallen to its lowest market capitalization since early 2019, before the generative AI boom propelled the chipmaker to the top of Wall Street. Since peaking on May 14, shares have dr... Read original →

NVIDIA’s valuation pullback reflects a recalibration of AI investment logic, not weakening fundamentals. Technically, high-bandwidth memory (HBM) has become the AI training bottleneck—Micron’s 84.9% gross margin signals its shift from peripheral to pivotal, forcing GPU architectures toward tighter compute-memory integration. On compliance, tightening U.S. export controls on advanced chips are accelerating global supply chain diversification, benefiting Intel and AMD with mature-node capacity and localized manufacturing. Strategically, AMD is scaling cloud adoption of MI300 accelerators, while Intel pivots to advanced packaging and foundry services to bypass GPU saturation. Over the next 12–24 months, capital will rotate from pure compute obsession to full-stack ‘compute-memory-interconnect’ exposure, making memory and packaging the new valuation benchmarks—punishing overreliance on single-customer dominance.