Nvidia May Be the Most Valuable Company in the World, but It Looks Like a Bargain Compared to These 3 Artificial Intelligence (AI) Stocks

Nvidia is framed as not overvalued despite a ~$5T market cap, with a P/E of 33 vs the S&P 500’s 26 and >950% total returns over five years. By contrast, the article flags Advanced Micro Devices as very expensive at a P/E of 188 even after a 38% revenue rise to $10.3B (period ended March 28), noting the stock is up ~160% this year. It warns that if AMD doesn’t deliver stronger results than its latest quarter, the stock could be vulnerable to a sell-off.

Analysis

This is not a broad AI sell signal; it is a relative-value warning that the market is willing to pay up for scarcity only when earnings quality and forward visibility are exceptional. On that basis, NVDA should remain the “safe” AI beta: it has the cleanest monetization path, the least dependence on a single second-order use case, and the strongest ability to defend premium multiples if hyperscaler capex stays firm. The article is more negative for the higher-beta beneficiaries of AI spend, where expectations have outrun evidence.

AVGO and AMD are the key losers because their upside is more sensitive to continued customer willingness to diversify away from NVDA. If AI budgets are scrutinized, custom silicon and alternative accelerator stories get marked down first: the market will question whether incremental share gains come with enough margin durability to justify triple-digit or near-triple-digit multiples. PLTR is different but still vulnerable; at its valuation, even a small deceleration in net new bookings can compress the multiple faster than fundamentals deteriorate.

The catalyst path is mostly 1-3 months: hyperscaler earnings, capex commentary, and any evidence that AI ROI is being measured more tightly. Over 6-18 months, the real falsifier is whether AI demand broadens beyond a handful of buyers; if it does, AVGO’s custom-chip thesis improves and the relative case for NVDA narrows. For now, the consensus appears to be missing that the trade is not “AI versus no AI,” but “quality AI compounder versus crowded second-tier AI duration.”

Key Decisions for Investors

• Long NVDA / short AMD in a 1-3 month relative-value pair. Rationale: NVDA should preserve premium quality and pricing power, while AMD is most exposed to disappointment versus already-lofty expectations. Falsify if AMD guides materially above consensus or wins a major share-of-wallet headline.
• Trim or hedge AVGO on strength ahead of hyperscaler earnings. Use a call overwrite or partial short against long-only exposure if you need to keep the position. Risk/reward favors downside if AI capex chatter softens; cover if AI revenue mix keeps accelerating above prior run-rate.
• Avoid initiating new PLTR longs at current valuation without evidence of reacceleration in commercial growth. If already long, use a tight earnings-tape stop over the next quarter; the stock can de-rate 20-30% on merely good-not-great guidance.
• Watch for a post-earnings rotation into NVDA versus the rest of the AI basket. If NVDA outperforms on a clean beat while AVGO/AMD/PLTR lag, add to the pair for a 4-8 week trade.
• If you want pure sector exposure, prefer SMH over single-name AMD/PLTR until capex visibility improves. The ETF dampens idiosyncratic multiple compression while keeping exposure to the AI spend cycle.