Berkshire Hathaway Has a New Top Holding, and It Dominates Its Market

Berkshire Hathaway increased its Alphabet stake by 45% in Q2, adding 24.5 million shares and bringing its holding to about 79 million shares worth $28.2 billion, or 9.41% of its portfolio and its fourth-largest position. Alphabet delivered Q2 revenue growth of 24% to $120 billion and operating-income growth of 30% to roughly $41 billion, but raised 2026 capex guidance to $195 billion-$205 billion and posted a $5.8 billion free-cash-flow outflow, its first negative FCF quarter as a public company. Shares have declined about 7% over the past month amid AI-spending concerns, leaving the stock valued at roughly 16x earnings.

Analysis

The relevant signal is not Berkshire’s ownership itself—13F disclosures are backward-looking and do not establish an ongoing bid—but the implied capital-allocation regime at Alphabet: near-term cash conversion is being exchanged for capacity that can protect Cloud growth and defend Search distribution. The market will reward this only if incremental infrastructure spend produces sustained Cloud backlog conversion and stable consolidated operating margins; otherwise, the valuation discount can persist despite apparently low headline earnings multiples.

Second-order beneficiaries are AI infrastructure vendors such as NVDA, AVGO, ANET and VRT, but Alphabet’s spending adds to an already crowded hyperscaler-capex narrative rather than creating a new one. The more differentiated implication is that capacity constraints may shift competitive share in enterprise AI: if Alphabet converts capacity into faster Cloud growth, it pressures AMZN and Microsoft’s cloud pricing/contract economics; if demand is weaker than management claims, all hyperscalers face simultaneous depreciation and power-cost leverage, making margins the common downside channel.

Over the next 1-3 months, the key catalyst is evidence that Cloud revenue acceleration and backlog support the higher run-rate spend. Over 6-18 months, the debate shifts from capex magnitude to return on invested capital: depreciation, power procurement, and networking costs will increasingly determine whether AI monetization expands or compresses earnings quality. Contrarian view: the market may be over-penalizing a single quarter of weak cash conversion, but it is underpricing the risk that Search monetization—not Cloud—ultimately funds the AI buildout and remains exposed to product disruption and regulatory remedies.

Key Decisions for Investors

• Accumulate GOOG in tranches ahead of the next earnings print only if Cloud growth re-accelerates and management maintains operating-margin discipline; target a 6-12 month holding period. Thesis fails if Cloud growth decelerates despite elevated capex or if forward operating-margin guidance is cut.
• Use a 3-6 month pair: long GOOG / short AMZN in equal beta-adjusted dollars if Google Cloud shows sequential growth-share gains. This isolates relative AI-capex monetization; exit if AWS re-accelerates materially or Alphabet’s Cloud margin/backlog deteriorates.
• Do not chase NVDA, AVGO, ANET or VRT solely on this spending signal. Maintain alerts for supplier guidance tied specifically to Alphabet deployments; a capex increase without raised supplier revenue outlook would indicate longer lead times or internally constrained buildouts rather than incremental near-term demand.
• For BRK.A/BRK.B, treat the disclosed Alphabet position as modestly supportive of portfolio NAV but not an investment catalyst. The actionable monitor is whether Berkshire continues rotating away from lower-growth cyclicals such as CVX; absent subsequent filings or repurchase data, there is no clean directional trade.