A Reddit StockMarket post has correctly flagged that more than half of Alphabet and Amazon's reported Q1 earnings came from unrealized gains tied to their Anthropic stakes, turning private AI valuations into a direct line item on trillion-dollar public-company income statements and raising the question of how much of Big Tech's profitability is now a function of startup paper gains rather than core operations.
The accounting is straightforward but the implications are not. Alphabet and Amazon hold strategic stakes in Anthropic, with Alphabet owning about 13 percent and Amazon holding a similar position through AWS. Those stakes are marked to market each quarter, meaning the companies estimate the private valuation of Anthropic and record an unrealized gain or loss based on the change from the prior period. In Q1 2026, Anthropic's valuation was marked up substantially, and that mark-up flowed through as a non-cash gain on the income statement. The Reddit post noted that those gains accounted for more than half of the headline net income figures for both companies, which is a material enough contribution to matter for anyone trying to understand what the earnings actually say about the underlying business.
That is a new phenomenon in public company earnings. Big Tech has always held stakes in private companies, but those stakes were rarely large enough relative to the parent company's scale to move the needle on headline profitability. The AI cycle has changed that dynamic because the valuations of frontier AI labs have grown so quickly that they now represent a meaningful percentage of the hyperscaler's balance sheet. Alphabet's Anthropic stake alone is worth more than $10 billion at recent private valuations, and Amazon's is similar. When those valuations move up 20 or 30 percent in a single quarter, the resulting gain is hundreds of millions of dollars that lands directly on the income statement. That is not revenue from cloud contracts or search ads. It is a valuation adjustment on a private investment.
The feedback loop is the part that makes this interesting. Hyperscalers are investing in AI labs not just for strategic alignment but because those labs are the largest customers of their cloud infrastructure. Anthropic runs much of its training and inference on AWS and Google Cloud. The cloud revenue those workloads generate is real operating profit. But the valuation gains that flow from those same relationships are paper profit. The result is a virtuous cycle where cloud spend from the AI lab lifts its valuation, which lifts the hyperscaler's earnings, which supports more cloud investment, which supports higher valuations. That is a reasonable dynamic, but it is also a dynamic that can reverse if the AI lab valuations correct or if the cloud contracts underperform. Investors need to be able to separate the operating reality from the accounting feedback loop.
That separation is harder than it looks. Headline earnings numbers are what analysts and investors react to first, and those numbers now include material contributions from private AI stakes. Alphabet's Q1 net income was $23.7 billion, with the Anthropic gain accounting for roughly 60 percent of that according to the Reddit analysis. Amazon's numbers followed a similar pattern. Those contributions make the headline numbers look stronger than the core business performance alone would justify. The risk is that the market begins to price the hyperscalers on the assumption that AI startup gains will continue to flatter earnings indefinitely, when the reality is that those gains are volatile, illiquid, and dependent on private market sentiment.
For San Francisco founders and investors, the dynamic has direct implications. AI startup valuations are no longer just a private market phenomenon. They are now material to the earnings of the companies most likely to acquire those startups. That creates a tighter coupling between private valuations and public market performance, which can amplify both upside and downside. A correction in AI lab valuations would not just hurt the venture ecosystem. It would show up in the next quarter's earnings from Alphabet and Amazon, which would create pressure on cloud pricing and deal terms. Conversely, continued valuation gains create a tailwind for the hyperscalers that makes them more willing to invest aggressively in the AI labs driving those gains. The result is a market where private AI valuations have public market consequences, and where the feedback loop between cloud spend and startup valuations is becoming a core feature of the ecosystem.
The practical lesson is about how to read those earnings. Investors should treat the AI stake gains as a separate line item, not as part of core operations. They should model the hyperscalers with and without those contributions to understand the underlying business performance. Founders raising from or selling to those hyperscalers should understand that their valuation affects not just their own cap table but the public earnings of their largest customers. That is a new kind of alignment, and it is one that the market is still learning how to price.
Also read: Match Group is slowing hiring to fund AI and consumer internet has its new margin script • Amazon's e-book market is getting flooded by AI-generated titles and the discoverability crisis is just beginning • Anthropic renting compute from Elon Musk's infrastructure shows how desperate the GPU scramble has become
