3 Artificial Intelligence (AI) Stocks That Are Expected to Grow at Faster Rates Than Nvidia Over the Next Couple of Years, According to Analysts

The article argues that Nvidia's growth may slow to a 26.2% CAGR through 2028, while Broadcom, AMD, and Marvell are projected to grow faster at 35.6%, 35.2%, and 30.3%, respectively. It highlights AI-driven demand for custom chips, Nvidia's $2 billion investment in Marvell, and Broadcom's potential to exceed $100 billion in chip revenue in 2027. The piece is largely comparative and forward-looking, with modestly positive implications for the named rivals but no immediate catalyst.

Analysis

The market is starting to price a shift from GPU scarcity to ecosystem fragmentation. That is not necessarily bearish for AI capex overall, but it is a relative negative for NVDA because every incremental dollar of hyperscaler spend that migrates into custom ASICs or alternative accelerators comes with lower attach rates in networking, software stack lock-in, and refresh-cycle control. The second-order winner is not just AVGO/MRVL/AMD, but also the broader custom silicon supply chain: EDA, advanced packaging, and foundry capacity should remain tight even if headline AI spend rotates away from Nvidia-branded systems.

The key nuance is timing. Over the next 2-4 quarters, these names can continue to rerate on order visibility and design-win announcements even before revenue inflects, while NVDA likely remains a higher-quality compounder with slower marginal upside. The risk is that consensus is extrapolating every design win into linear TAM capture; in practice, custom silicon ramps are lumpy, and many programs underdeliver on yield, software readiness, or deployment scale. That makes the setup better for tactical trades than outright long-only beta.

Contrarian view: the market may be underestimating how much custom chip adoption still depends on Nvidia’s ecosystem. Nvidia can lose share in specific workloads while still expanding overall wallet share through networking, interconnect, and software monetization. In that sense, the most likely near-term loser is not NVDA earnings growth per se, but sentiment multiple expansion, especially if hyperscaler capex guidance normalizes and investors start demanding proof of monetization rather than just AI capacity buildouts.

Key Decisions for Investors

• Relative value: long AVGO / short NVDA for 3-6 months. AVGO has the cleanest direct exposure to hyperscaler ASIC substitution; use the pair to isolate share-shift upside while reducing broad AI-beta risk.
• Tactically long AMD on any post-launch weakness over the next 1-2 quarters. The stock is likely to trade on MI450 cadence and partnership headlines before the street fully models revenue contribution; target a 1.5-2.0x upside versus a tight stop if launch adoption slips.
• Add MRVL as a smaller satellite long, but only on pullbacks. It is the highest beta way to express custom-chip penetration, yet also the most execution-sensitive; size accordingly because any design-win disappointment could compress the multiple quickly.
• Hedge NVDA exposure with downside structures into the next earnings cycle, e.g. put spreads 1-2 quarters out. The thesis is not a crash, but a period of slower multiple expansion as growth decelerates from peak rates.
• Monitor GOOGL and META capex commentary as the leading indicator for the trade. If hyperscaler AI spend shifts toward internally designed silicon, increase AVGO/MRVL exposure; if spend re-accelerates broadly, cover NVDA hedges and rotate back to the highest-quality platform name.