Billionaire David Tepper of Appaloosa Is Piling Into the AI Revolution With These 3 Big Purchases

David Tepper’s Appaloosa made major AI-related purchases in the March quarter, nearly doubling Amazon, more than tripling Uber, and initiating Sandisk. The article argues all three have durable advantages and double-digit growth potential, with AI boosting AWS, Uber’s routing/pricing, and Sandisk’s NAND storage demand for AI data centers. The piece is largely bullish commentary rather than a new company-specific catalyst, so near-term price impact is likely limited.

Analysis

This is less a generic AI-basket trade than a differentiated signal that capital is rotating toward the monetization layer of AI rather than the model layer. The market is increasingly rewarding companies that can absorb AI into existing distribution, which means the upside is not just incremental revenue but higher utilization of fixed assets, better pricing power, and margin expansion from software-like economics. That makes AMZN and UBER particularly interesting because AI can improve unit economics without requiring a proportional jump in capex intensity.

The second-order implication is that the real squeeze is moving upstream into memory and storage, where supply is structurally tighter than most investors assume. If enterprise AI deployment continues to favor persistent data retention, low-latency retrieval, and ever-growing inference logs, NAND pricing can stay elevated longer than consensus expects, which would extend SNDK’s earnings inflection well beyond a normal cyclical peak. The risk is that the market is extrapolating a shortage regime into a durable supercycle; any incremental supply, weaker hyperscaler capex, or a shift toward lower-storage architectures would compress the multiple fast.

The contrarian read is that the move into AI-adjacent names may already be crowding into the most visible beneficiaries while underpricing execution risk in the applications stack. For AMZN, the concern is that AI features can drive usage but not necessarily monetization if competitive pass-through forces remain tight in cloud and retail. For UBER, the market may be overestimating how much AI can lift take rates versus simply lowering service costs; the opportunity is real, but the cleaner P&L leverage probably shows up over multiple quarters, not days.

In the near term, sentiment can stay self-reinforcing because these names offer a cleaner way to express AI exposure than the crowded semiconductor leaders. The better setup is to own the cash-flow compounders into confirmation, while fading any parabolic move in SNDK if forward estimates stop accelerating. If AI capex cools, SNDK will likely be the first to re-rate lower; if AI adoption broadens, AMZN and UBER should compound more steadily with less multiple fragility.

Key Decisions for Investors

• Overweight AMZN on a 3-6 month horizon: use pullbacks to add exposure ahead of the next AWS/advertising margin revision cycle; best risk/reward is via call spreads rather than outright stock if implied vol remains contained.
• Initiate a long UBER / short high-multiple mobility basket trade for 1-2 quarters: UBER has a clearer path to AI-driven operating leverage, while weaker platform peers are more exposed to pricing pressure and subsidy creep.
• Treat SNDK as a tactical momentum long only, not a core holding: pair long SNDK against short NVDA or a semiconductor ETF if you want to isolate storage scarcity from broader AI exuberance; take profits aggressively on any estimate stalling.
• For new capital, prefer AMZN over SNDK at current levels: AMZN offers AI upside with materially lower earnings-duration risk, while SNDK is most vulnerable to any normalization in NAND pricing within 6-12 months.
• Watch for a 1-2 quarter confirmation window: if cloud growth and rideshare booking trends do not inflect meaningfully by then, reduce AI-exposure baskets and rotate into less crowded cash-flow names.