31% of Berkshire Hathaway's Portfolio Is Riding on These 2 AI Stocks Under Greg Abel

Berkshire Hathaway has roughly 31% of its equity portfolio invested in AI-linked holdings, with Alphabet representing just under 10% after Berkshire acquired nearly 106 million shares and Apple accounting for just over 20%. Alphabet's revenue growth accelerated to 24% over the last year and Google Cloud growth reached 82%, while it generated $53B in free cash flow despite projected annual AI-related capex of $195B-$205B. Apple remains a major Berkshire holding despite trading at 38x earnings, supported by 16% year-over-year fiscal Q3 2026 revenue growth and its Gemini AI partnership with Alphabet.

Analysis

The investable implication is less Berkshire's AI exposure than a potential validation of Alphabet as the lowest-multiple hyperscaler with a credible path from AI capex to monetization. If Cloud growth and search monetization continue to absorb elevated infrastructure spend, GOOGL can sustain earnings revisions while its multiple remains below MSFT and NVDA; that combination is more powerful than a generic AI ownership narrative. The key near-term issue is whether incremental AI revenue converts to gross-profit dollars faster than depreciation and power costs rise over the next 2-4 quarters.

Apple's reliance on an external foundation-model partner shifts part of the AI value pool toward the model and cloud provider rather than the device vendor. AAPL can still benefit if AI features pull forward replacement cycles or support premium pricing, but absent measurable iPhone mix, services attach, or gross-margin uplift, its premium multiple leaves little room for a merely adequate AI rollout. The relevant 6-18 month question is whether Apple develops differentiated on-device capabilities and bargaining leverage, rather than whether it can advertise AI features.

BRK.A is an imperfect way to express this view: public-equity look-through exposure is diluted by insurance, railroad, utility, and cash-balance outcomes, while the market will not capitalize its portfolio at the same multiple as a pure-play owner. Any disclosure-driven buying signal is also inherently lagged. The contrarian point is that large Alphabet ownership may tighten the perceived valuation floor, but it does not eliminate the risk that hyperscaler capex becomes a lower-return arms race; a slowdown in Cloud backlog conversion would re-rate the entire thesis quickly.

Key Decisions for Investors

• Initiate a 3-6 month long GOOGL / short AAPL pair, sized beta-neutral, only if GOOGL maintains positive FY earnings revisions while AAPL forward estimates remain flat. Target 10-15% relative outperformance; exit if Alphabet reports decelerating Cloud growth or materially weaker free-cash-flow conversion.
• Prefer direct GOOGL exposure to BRK.A for AI-related capital deployment. Use BRK.A only as a lower-volatility holding where insurance underwriting, buybacks, and cash deployment are the intended return drivers; do not underwrite it as an AI proxy.
• Ahead of the next Alphabet earnings release, monitor Cloud revenue, capex guidance, depreciation growth, and management commentary on AI query monetization. If capex rises faster than revenue while operating margins compress, avoid adding despite a superficially low P/E.
• Do not add to AAPL solely on AI narrative momentum at the current premium valuation. Upgrade the view only after evidence of AI-driven unit mix, pricing, services attach, or margin expansion; a guidance cut to iPhone growth would invalidate the upgrade-cycle support case.