Nvidia Just Delivered Bad News for AMD and Intel

“Nvidia’s entry into the CPU market creates a structural margin headwind for Intel and AMD by forcing them to compete on performance-per-watt against a company that already owns the entire AI software stack.”

Nvidia’s pivot into the CPU market with the Grace/Vera architecture isn't just a product launch; it’s a vertical integration play that threatens the x86 duopoly of Intel and AMD. By optimizing CPUs specifically for agentic AI orchestration, Nvidia is capturing the 'glue' layer of the data center stack. If Nvidia hits that $20 billion CPU revenue target, they effectively commoditize the very chips that have been the lifeblood of AMD and Intel’s data center growth. However, the market is pricing in perfection, and any sign of supply chain friction or a slowdown in agentic AI adoption will lead to a violent de-rating of NVDA’s premium valuation.

“Vera CPU revenue projections are directionally bullish but rest on unvalidated assumptions about production adoption and competitive response; the $20B target should be treated as aspirational guidance, not a lock.”

Nvidia's $20B CPU revenue projection for FY2027 is material but needs scrutiny. The article conflates two separate dynamics: (1) genuine agentic AI demand for orchestration, and (2) Nvidia's ability to capture that demand. Vera is unproven in production at scale—1.8x performance claims are lab benchmarks, not field results. AMD and Intel have entrenched relationships, mature supply chains, and lower switching costs for legacy workloads. Nvidia's GPU dominance doesn't automatically translate to CPU markets where they're a new entrant. The 70% FY2028 revenue guidance is impressive but depends heavily on whether this CPU bet materializes—if Vera underperforms or adoption stalls, that guidance could compress significantly. Also missing: Vera's gross margins versus GPU margins, and whether $20B is additive or cannibalistic to GPU revenue.

“The 2028 70% growth target is an outsized bet that may hinge on unproven CPU monetization and sustained AI capex, making the upside vulnerable to demand normalization and margin compression.”

Biggest takeaway: Nvidia guiding to 70% revenue growth in fiscal 2028 vs Street models ~44% is an outsized bet; the article’s bullish read on Vera CPU and 'agentic AI' risks conflating product announcements with durable demand. The acceleration would require continued hyperscale capex, AI tooling adoption, and price/mix tailwinds, while AMD/Intel still chase CPU-less compute and margin pressure could intensify as Nvidia expands into CPUs. Disclosed data center revenue mixes, backlog visibility, and gross margin trajectory remain unclear, making the 2028 target highly sensitive to a few large orders and cycle timing.

“Nvidia's credible $20B CPU forecast for FY2027 introduces a new competitive vector that threatens to erode AMD and Intel data-center growth rates starting in 2026.”

Nvidia's $20B CPU revenue target for FY2027 directly challenges AMD and Intel in the agentic AI segment, where CPUs handle orchestration rather than raw inference. With Vera claiming 1.8x faster task completion and 5x bandwidth per watt, Nvidia could capture share in a market where AMD posted $6.7B and Intel $6.3B in Q2 data-center revenue. Yet the projection assumes rapid adoption of agentic workflows and flawless scaling of a brand-new architecture against entrenched x86 ecosystems. Execution delays, software compatibility gaps, or slower-than-expected agentic demand could blunt the impact, leaving AMD and Intel's existing server franchises largely intact through 2026.