Billionaire David Tepper Sold Nvidia Stock and Piled Into This AI Stock That's Up Nearly 200% in the Past 5 Years

Appaloosa founder David Tepper increased his Alphabet stake by 28.8% in Q4 while trimming Nvidia by ~10.5%, signalling a notable portfolio shift into Google parent shares. Alphabet shares have risen ~190% over five years versus a 75.7% gain for the S&P 500, driven by AI integrations (Gemini, search AI and chatbot), faster-growing cloud revenue (3rd largest IaaS share), expanding YouTube streaming/subscriptions, and long-term upside from Waymo. Article frames these diversified growth avenues and limited antitrust fallout as a strong buy-and-hold case for Alphabet tied to the AI thematic.

Analysis

Tepper’s reallocation from Nvidia into Alphabet signals a tactical rotation from hardware-driven, capex-dependent AI exposure to platform owners with higher gross margins and recurring monetization levers. Over the next 12–24 months, Alphabet’s optionality (LLM APIs, search-ad overlays, YouTube subscriptions and cloud elasticity) compounds revenue per DAU without linear compute cost growth — meaning incremental AI monetization can show margin expansion even if raw inference costs remain sticky.

Second-order winners include ad-tech partners (DSPs, verification vendors) and Google Cloud’s hyperscale partners who capture the long tail of enterprise LLM deployments; losers are smaller pure-play inference infra vendors if customers standardize on cloud-hosted models. Supply-chain effects: sustained LLM adoption that favors cloud-hosted inference reduces incremental GPU spot-market arbitrage and places an eventual cap on standalone inference hardware multiples over 2–4 quarters.

Key risks are regulatory shocks (privacy/antitrust remedies that alter ad targeting economics) and margin compression if large customers successfully negotiate fixed-price inference SLAs or pivot to cheaper open models — either could shave 5–15% off near-term advertising and cloud growth assumptions. Monitor three catalysts: quarterly ad RPM trends (next 2–3 quarters), enterprise LLM bookings in Google Cloud (next 4 quarters), and any antitrust enforcement milestones (12–36 months) that could force structural changes to distribution or pricing.

Key Decisions for Investors

• Long core: Initiate a 6–12 month long position in GOOGL (or GOOG) sizing 2–4% of portfolio. Target: 20–30% upside if ad RPMs and Cloud LLM ARR accelerate; downside: 10–15% on macro ad slowdown. Hedge: buy 6–12 month 10–15% OTM puts for 150–300 bps cost if volatility cheapens.
• Pair trade (directional rotation): Long GOOGL vs short NVDA execution — 1:0.3 notional (i.e., $1 of GOOGL exposure funded by $0.3 NVDA short) over 9–12 months. Rationale: capture platform monetization re-rating while capping exposure to cyclical GPU multiple contraction; risk: NVDA outperformance if hardware demand surprises upwards; set stop-loss at 12% on the short leg.
• Event-driven optionality: Buy a 12–24 month call spread on GOOGL funded by selling short-dated calls (calendar) to take advantage of slower IV roll-off. Structure: buy 12–24m ITM/ATM call and sell 3–6m OTM call; target asymmetric payoff with net debit ~3–6% of notional and upside capped but >3x on net debit if multi-quarter re-acceleration occurs.
• Contrarian hedge: If you prefer neutral exposure, sell a small amount of near-term (3–6m) call overwrites on existing GOOGL stock to collect premium and buy 18–24m out-of-the-money puts as tail protection. This reduces carry cost while protecting against a regulatory or demand shock that could drop shares >20%.