Why this $1.5-billion money manager is buying Amazon and selling gold

Richard Croft says investors are in a "serious bull market" and sees the S&P 500 reaching 8,000 by year-end, citing AI spending and U.S. tax cuts as key catalysts. He is overweight technology and financials, with Amazon at about 10.5% of the portfolio and AMD at 5.6%, while Circle Internet Group is a 2% position tied to stablecoins and prediction markets. Croft trimmed gold exposure after the rally, selling GLD and reducing Agnico Eagle, reflecting a shift in demand and profit-taking.

Analysis

The market setup is less about broad “risk-on” and more about a narrowing dispersion trade: AI capex plus easier tax policy is extending duration for the winners with the strongest operating leverage to secular spend. That is constructive for platform winners with adjacent monetization layers, but it also raises the bar for any company whose multiple already discounts flawless execution. In that context, the most important second-order effect is not just higher revenue growth, but a continued re-rating of firms that can turn incremental AI demand into recurring margin expansion rather than one-time hardware sales.

Amazon stands out because it gives investors three shots on goal: retail operating efficiency, cloud re-acceleration, and AI adjacency. The market still tends to value it as a lower-quality mega-cap than the top-tier software compounders, which creates room for multiple expansion if AWS growth stabilizes while retail margin mix improves. By contrast, AMD is the cleaner expression of AI spending convexity, but the trade becomes increasingly dependent on share gains and supply discipline rather than simply riding industry-wide demand; that makes it a good relative long versus more saturated AI bellwethers.

CRCL is a different type of winner: it is not about near-term fundamentals so much as embedded optionality on digital settlement infrastructure and prediction-market adoption. The risk is that this narrative can outrun actual monetization, making the stock highly sensitive to regulatory friction, volume concentration, and sentiment reversals in crypto-linked assets. That said, if stablecoins become a core rails layer for transactional finance, the upside is asymmetric because the addressable market expands from crypto-native use cases into broader payment flows.

Gold looks like the clearest positioning unwind rather than a macro call. If central-bank demand has peaked and the dollar remains the main macro hedge, gold’s marginal buyer weakens just as consensus is most crowded. That sets up a rotation from “store of value” into cash-generative growth, but the main risk to the bullish tape is a sharp rates-up / growth-down regime shift that would punish both long-duration tech and speculative digital asset exposure at the same time.

Key Decisions for Investors

• Add AMZN on pullbacks over the next 2-4 weeks; target a 6-12 month hold for multiple expansion if AWS growth inflects. Risk/reward is favorable because downside is cushioned by operating leverage in retail, while upside comes from re-rating toward other AI beneficiaries.
• Go long AMD vs. short NVDA as a relative-value pair for 3-6 months. The trade expresses AI capex persistence while avoiding the law-of-large-numbers risk embedded in the larger platform winner; use a 1:1 dollar-neutral structure and exit if AMD loses share momentum.
• Take a small tactical long in CRCL only via call spreads or a starter equity position, sized for binary regulatory/market-structure risk. Horizon is 6-18 months; this is an optionality trade, not a core compounder position.
• Reduce or avoid new gold exposure until the next central-bank reserve data confirms renewed accumulation. For existing holders, consider selling upside via covered calls over 1-2 months to monetize elevated sentiment and protect against a continued drift lower.
• Pair long AMZN / short GLD for a 3-6 month macro-neutral expression of the rotation from defensive stores of value into AI-enabled growth. This works best if rates stay range-bound and fiscal/AI tailwinds continue to dominate risk appetite.