29.7% of Berkshire Hathaway's $356 Billion Portfolio Is Invested in 2 Artificial Intelligence Stocks

Berkshire Hathaway’s new CEO Greg Abel has quadrupled Berkshire’s stake in Alphabet since the start of 2026, with Alphabet now at 78.8M Class A and 27.2M Class C shares worth $36.6B (10.2% of the portfolio). Alphabet reported Q2 2026 revenue of $63.3B (+17% YoY) alongside 1B+ monthly active users for AI Mode and a record $514B Google Cloud order backlog, while Google Cloud revenue surged 82% to $24.8B. Apple is the largest remaining single holding (19.5%), supported by its Apple Intelligence push—though Alphabet and Google Cloud are the clearer near-term AI growth drivers for Berkshire’s portfolio.

Analysis

The market should treat this less as an “AI winner” revelation and more as a signal that Berkshire’s capital is still gravitating toward two high-quality cash compounding machines. That matters because both names already trade on durability; incremental institutional endorsement can tighten the discount rate, but only if the next few quarters show AI is accretive without forcing a step-up in capex that bleeds free cash flow.

Alphabet has the cleaner near-term setup. The real debate is not whether AI features attract usage, but whether they preserve monetization per query while keeping cloud investment from outrunning operating leverage. If that equation holds, the stock can keep re-rating; if not, the market will start valuing it more like a capital-intensive infrastructure business than a software platform. Apple’s AI angle is slower-burn: the upside is a replacement-cycle lift and ecosystem stickiness, not immediate AI revenue, so the stock likely needs hardware evidence before it deserves a higher multiple.

The contrarian miss is that consensus is focusing on “AI exposure” while underweighting timing. Alphabet has a 1-3 quarter catalyst path; Apple’s payoff is more like 6-18 months and hinges on convincing consumers to upgrade. On the downside, the most important falsifier for Alphabet is any sign that AI search cannibalizes paid clicks or that capex guidance steps up faster than revenue; for Apple, the falsifier is another cycle where AI features feel incremental rather than essential.

Key Decisions for Investors

• Go long GOOGL on 3-5% pullbacks ahead of the next earnings window; thesis is that monetization durability beats capex anxiety. Falsify if search revenue growth decelerates materially or capex guidance steps up without a matching cloud backlog conversion.
• Pair trade: long GOOGL / short AAPL for 3-6 months. Alphabet has the nearer-term AI monetization path; Apple’s AI upside is more dependent on a later hardware refresh cycle. Cover if Apple intelligence drives a visible upgrade-cycle inflection.
• Use BRK.B as a low-beta quality basket only on weakness, not as a momentum trade. The stock benefits from the market re-pricing Berkshire as a disciplined allocator of excess cash into durable AI-enabled compounders, but the upside is likely slower than the underlying holdings.
• Set an alert on GOOG/GOOGL if free cash flow margin compresses meaningfully while revenue growth remains strong; that would signal the AI buildout is becoming value-destructive rather than defensive.