Berkshire Trimmed Amazon to 2.3 Million Shares. Why That May Be a $300 Billion Mistake.

Berkshire's trim may be old information, not a verdict on Amazon

Berkshire Hathaway cut its Amazon stake, and Amazon later added roughly $300 billion in market value. That gap is the core risk for investors who treat Berkshire's filing as a current read on Amazon's thesis.

The 13F shows what Berkshire held, not what changed after earnings

Berkshire's disclosure was its Q1 2026 13F filed May 15 for the quarter ending March 31, and it showed Amazon reduced to roughly 2.3 million shares. That tells you what had been sold by late spring. It does not capture the market's later reaction to Amazon's cloud acceleration and AI demand.

Berkshire's moves look more like rotation than a broken-moat call

The broader portfolio overhaul looks less like panic selling and more like heavy rebalancing. Greg Abel's first quarter included a 224% increase in Alphabet, along with new positions in Delta Air Lines and Macy's, as part of one of Berkshire's biggest portfolio renovations ever. That fits a broader pattern of trimming winners, reducing concentration, and redeploying capital elsewhere.

Why copying the trim may have been expensive

By the time Amazon's latest results hit, the market was rewarding the company for stronger cloud growth, not just future promise. Shares jumped more than 12% before the bell after the company posted its strongest cloud growth in over four years, putting it on track to add about $300 billion in market value.

So the real question is not whether Berkshire made a legal trade. It is whether Berkshire exited before Amazon's earnings engine started accelerating again.

Amazon's operating engine is still widening

The key question is no longer whether demand exists. It is whether Amazon is turning that demand into broader profit power. The latest numbers suggest it is.

Q1 showed growth plus stronger margins

In Q1, revenue rose 17% to $181.5 billion. More important, AWS grew 28%, generated $14.2 billion in operating income, and did it at a 37.7% operating margin. That matters because AWS is not just a revenue line anymore; it is becoming an increasingly large part of Amazon's earnings base.

AWS acceleration does not look like a one-quarter outlier

Management said AWS posted its fastest growth rate in 15 quarters and had the largest Q4-to-Q1 AWS revenue increase ever. For a business of that size, sustained acceleration usually forces investors to reassess the whole company, not just the cloud segment.

Amazon's AI spend is starting to show customer proof

Management also said AWS's AI revenue run rate is over $15 billion. That does not prove every dollar of capex will be rewarded immediately, but it does suggest customer demand is already helping offset the investment case. When spending of that kind is backed by real revenue, investors are more likely to view the cycle as constructive than purely speculative.

The bear case is simple: capex may outrun monetization

Bears are not wrong to focus on the bill coming due. Amazon's free cash flow decreased to $1.2 billion on a trailing 12-month basis, while capital expenditures reached $44.2 billion in the quarter. Reuters also reported management said demand was so strong that capacity remained insufficient to serve customers.

That is the real trade-off. Amazon is front-loading infrastructure in anticipation of AI demand. If that demand fades, the spending can weigh heavily on cash generation. If demand holds, today's investments are more likely to look like capacity for a wider profit engine than wasted scale.

What matters now is cash conversion, not Berkshire's old filing

Berkshire's 13F matters less than whether Amazon can turn rising spending into durable returns. Amazon is still roughly 16.84% below the consensus 12-month price target, but the stock now has to justify a much larger capex commitment: $220 billion in planned capital expenditure.

The bullish trigger is monetary

The clearest positive sign would be evidence that the AI buildout is converting into cash, not just bigger revenue headlines. AWS's current revenue contribution helps, but stronger operating leverage and cash generation are what would make the trim look earliest.

What would invalidate the setup

The bearish break is straightforward: spending rises, but monetization does not follow. If capex keeps climbing toward $220 billion and cash generation stays near $1.2 billion on a trailing 12-month basis, investors are less likely to forgive the spending spree.

My read: Berkshire's Amazon trim may yet look premature. The filing was a positioning signal from months ago, not a final verdict on Amazon's cash-flow cycle. Until spending stops producing measurable returns, the market has reason to keep treating Amazon as an infrastructure winner rather than an efficiency problem.