3 Brilliant Growth Stocks to Buy Now and Hold for the Long Term -- Including Nvidia, of Course

The article is broadly constructive on Nvidia, Broadcom, and Arista Networks, highlighting strong AI-driven growth, attractive long-term fundamentals, and ongoing capital returns. Nvidia reported revenue up 85% year over year and raised its dividend by 2,400%, Broadcom posted $19.3 billion in quarterly revenue with AI revenue above $8.4 billion, and Arista delivered Q1 revenue of $2.7 billion, up 35%. The piece is primarily investment commentary rather than fresh news, so market impact is likely limited.

Analysis

The important second-order takeaway is that the AI capex cycle is broadening from compute into the networking and custom silicon layers, which tends to extend the duration of the trade rather than just rotate it. That favors NVDA, AVGO, and ANET in different ways: NVDA captures the “picks and shovels” premium at the GPU layer, AVGO monetizes design wins and software attachment, and ANET becomes the throughput bottleneck beneficiary as cluster sizes scale. If this phase persists, the market is likely underpricing the persistence of non-GPU spend, which can support multiple expansion even if headline GPU growth normalizes.

The clearest risk is not valuation in isolation but digestion risk: these are now crowded consensus winners, so any evidence of order pushouts, supply relief, or customer capex pauses can trigger sharp 10-15% de-ratings in days rather than quarters. ANET appears most exposed to near-term noise because supply constraints can create a false negative on demand, while AVGO may prove the most resilient because its software and custom ASIC mix dampens pure hardware cyclicality. NVDA remains the highest-beta expression; any deceleration in revenue growth or China-related uncertainty would likely compress the premium fastest.

Contrarian view: the article’s “undervalued” framing may be too simplistic because the real variable is not current P/E, but the sustainability of growth rates as the market embeds ever-higher terminal revenue assumptions. In that context, AVGO may be the better risk-adjusted long than NVDA because it has multiple monetization vectors and less dependence on one product cycle. The market is also likely underestimating the spillover winners outside the named trio—especially networking, optical, and memory bandwidth suppliers—because the more AI clusters scale, the more incremental value migrates away from accelerators themselves.

Key Decisions for Investors

• Stay long NVDA as the purest beta to AI capex, but prefer a staggered entry on 2-4 week pullbacks; risk/reward is attractive only if held through volatility, with 10-15% drawdowns likely on any guidance hiccup.
• Add AVGO on weakness versus NVDA as the better diversified AI monetization vehicle; 6-12 month upside should be more durable if custom silicon adoption keeps expanding, while downside is cushioned by software revenue.
• Initiate a relative-value long ANET / short a basket of slower enterprise networking names for a 3-6 month horizon; if AI networking intensity keeps rising, ANET should outperform as cluster throughput requirements tighten.
• Use call spreads rather than outright longs in NVDA and ANET if entering now; implied volatility is likely elevated, and spreads preserve upside while limiting multiple-compression risk over the next 1-3 months.
• Monitor for supply-chain relief or capex commentary from hyperscalers; if those emerge, reduce exposure quickly because the first reaction will likely be a factor unwind in crowded AI winners.