This Hedge Fund Just Sold Nvidia and Bought AMD Stock. Which Artificial Intelligence (AI) Chipmaker Is the Better Buy?

Tiger Global (Chase Coleman) initiated/added AMD to ~1.6% of its portfolio (~$400M) while reducing Nvidia by ~7% in Q2, leaving Nvidia as a ~9.3% position (~$2.2B). Despite AMD being up 110%+ in 2026 vs Nvidia up ~12%, the article argues Nvidia offers better value on a forward earnings basis because AMD trades at a higher P/E premium. Overall, the piece leans toward Nvidia as the preferable stock pick over the next few years, even if AMD outperforms in 2026.

Analysis

The market is increasingly treating this as a momentum-vs-valuation trade, not a pure fundamentals call. If large-cap growth allocators rotate from NVDA into AMD, the first-order effect is a near-term support bid for AMD from incremental ownership and a modest overhang on NVDA from narrative drift, but the second-order effect is more interesting: AMD’s higher multiple makes it more vulnerable to any deceleration in AI accelerator share gains, while NVDA’s lower embedded expectation gives it more room to re-rate on even average execution.

The key asymmetry is that hedge-fund flow can temporarily dominate price action in AMD, but it does not change the burden of proof on earnings quality. Over the next 1-3 months, AMD likely trades as a sentiment vehicle tied to AI spending prints and 13F imitation; over 6-18 months, NVDA should benefit if capital intensity in AI remains elevated and buyers continue prioritizing platform breadth, software attach, and supply assurance over incremental share gains. If AI capex slows or enterprise monetization disappoints, AMD’s multiple is where compression would hit hardest.

The contrarian point is that “cheaper” often means “less loved and less crowded,” not automatically “better value.” Consensus may be underestimating how much of AMD’s recent relative strength is already financing in optimistic share-take assumptions, while underweighting the optionality in NVDA if earnings simply remain robust. The trade to avoid is treating a 13F change as a durable thesis without confirming actual revenue mix, gross margin, and forward guide inflections.

The clean falsifier is a turn in forward AI spending commentary or any sign that AMD’s growth is plateauing while NVDA’s guide stays intact; that would likely unwind the relative multiple spread quickly. Absent that, this looks more like a tactical pair than a standalone long in either name.

Key Decisions for Investors

• Initiate a tactical pair: long NVDA / short AMD into the next earnings cycle; target 2-4% relative outperformance for NVDA over 4-8 weeks if AI capex commentary remains firm. Stop if AMD re-accelerates guide revisions or NVDA loses share in disclosed customer checks.
• For portfolios already long semis, rotate beta exposure toward NVDA via SMH call spreads rather than outright AMD longs; the risk/reward is better because NVDA has lower multiple compression risk if sentiment cools.
• Use AMD as a sentiment-trade, not a core hold: only add on post-earnings pullbacks if management raises near-term AI revenue guidance; otherwise the stock is vulnerable to 10-15% multiple compression over 1-3 months.
• Watch 13F-driven inflows into AMD and hedge them with NVDA on strength; if AMD outperforms on flow alone while fundamentals are unchanged, fade the move.
• Alert on next quarterly guide for both names: if NVDA sustains gross margin and data-center growth while AMD merely meets, expect relative valuation convergence in favor of NVDA over 6-12 months.