Access all our platforms from a single account.
From Apple to Alphabet: Warren Buffett’s late pivot into AI infrastructure
Key takeaways:
Berkshire’s Alphabet stake signals a reversal of the old Google “mistake” and reflects a belief that future AI profits concentrate in strong, cash-rich platforms
Strong cloud momentum, heavy investment in data centres and chips, and broad revenue streams turn Alphabet into a stable AI infrastructure play.
Trimming Apple while adding Alphabet suggests a preference for AI embedded in cloud, search and software ecosystems rather than tied to hardware cycles.
For years, Warren Buffett’s Google story was a cautionary tale. He openly admitted that not buying the stock was a major mistake, despite watching it turn online search into an advertising machine. Now, in the final months before he hands the chief executive role to Greg Abel, Berkshire Hathaway has quietly bought a multi-billion dollar stake in Alphabet, Google’s parent.
Alphabet’s Class C shares closed on 17 November at 285.60 USD, up 3.11% on the day, after news of the stake helped push the stock to record territory. The shares are up about 50% so far in 2025 and are the best performer among the so-called “Magnificent Seven” this year. When a value investor famous for avoiding fads buys into a market leader near all-time highs, the obvious question is: what does he see that the rest of us might be missing?
From Apples to Alphabets
Berkshire has been a net seller of equities for twelve straight quarters, including the latest one. It sold around 12.5 billion USD of stocks while buying about 6.4 billion USD, and allowed its cash pile to swell to a record 381.7 billion USD. That is not the behaviour of a man who thinks everything is cheap. Inside that overall selling, though, sits a big reshuffle.
Berkshire reduced its Apple stake by about 15% in the quarter and cut its position in Bank of America by around 6%. Alphabet, by contrast, arrives as the new kid in the top-ten club. Regulatory filings and portfolio breakdowns suggest the stake now ranks around tenth in Berkshire’s equity book, behind familiar names like American Express, Coca-Cola and Chevron. It is a rare tech-labelled position for a conglomerate that spent decades owning railroads, insurers and consumer staples, not high-growth software giants.
What changed is not so much Buffett’s principles as the companies themselves. Apple, which he has always described as a consumer brand, now lives in a world where hardware upgrades depend heavily on artificial intelligence features. At the same time, Alphabet looks less like a speculative tech name and more like a sprawling utility for the digital economy, with advertising and cloud revenues that look surprisingly steady for something built on lines of code.
Top 10 Holdings
Holding
% of Portfolio
Main Change
Apple
22.69%
Reduce
American Express
18.84%
—
Bank of America
10.96%
Reduce
Coca-Cola
9.92%
—
Chevron
7.09%
—
Occidental
4.68%
—
Moody’s
4.40%
—
Chubb Limited
3.31%
Add
Kraft Heinz
3.17%
—
Alphabet
1.62%
Buy
Source: Dataroma and Saxo Bank estimates
Alphabet as AI infrastructure, not a gadget story
Alphabet sits where AI ambition meets old-fashioned cash generation. In the third quarter of 2025 it reported around 102 billion USD of revenue, ahead of forecasts, with profits also beating expectations. The main driver was Google Cloud, which has shifted from “nice extra” to growth engine as AI firms rent its computing power.
On top of that, Alphabet’s Gemini models and AI-enhanced search now reach hundreds of millions of users. These tools run on a global network of data centres, custom chips and fibre that will require roughly 90-plus billion USD of investment spending this year. In simple terms, Alphabet wants to own more of the “picks and shovels” for the AI gold rush.
Its tie-up with Anthropic adds another layer. Google has invested billions in the start-up and agreed a large chip and cloud deal that should drive future workloads over Google Cloud. Berkshire’s stake gives it an indirect claim on that ecosystem: every Anthropic query that runs on Google hardware strengthens Alphabet’s role as AI infrastructure.
Crucially, this build-out sits on a strong balance sheet. Alphabet trades at about 25 times expected earnings, cheaper than some mega-cap peers, and still generates heavy free cash flow from search and YouTube. That cash can fund data centres and still support buybacks, which suits any investor who likes “wonderful companies at fair prices”.
What Buffett’s vote of confidence really signals
Buffett buying Alphabet is more than a simple seal of approval. It is a specific view on where AI profits will pool. Alphabet earns from search ads, YouTube, maps, app stores and cloud. AI is not a stand-alone product here. It is an upgrade that can deepen engagement and pricing across existing businesses.
It is also a tilt toward infrastructure over devices. Apple focuses on on-device intelligence but is still shaping its AI business model. Alphabet already earns from AI through cloud contracts, ad tools and productivity software. Trimming Apple while adding Alphabet hints that Berkshire sees more future AI value in data centres and platforms than in handset cycles.
Finally, it is a reminder that “tech” is not one bucket. Alphabet may share an index label with fast-growing AI start-ups, but its moat, cash generation and diversified revenues place it closer to the compounding machines Buffett has always favoured.
Risks that even Berkshire cannot wish away
The risks are real. Alphabet could overspend on AI capacity if customers slow projects or rivals win big deals. Watch Google Cloud growth, margins and long-term investment guidance. Regulation is another threat. Tougher antitrust or privacy rules in the United States or Europe could hit search and ad profits or force changes in how data is used.
Execution in AI also matters. Alphabet stumbled in the early rounds of generative AI and is still racing to regain the narrative with Gemini and other models. If users or corporate clients prefer rival tools, all that spending on chips and data centres could deliver lower-than-hoped returns, even with Buffett on the register.
Closing the loop: what Buffett’s Alphabet bet really teaches
Buffett spent years saying that not buying Google was one of his big mistakes. Alphabet sat in plain view, with search profits and cash flow, while he stayed on the sidelines. Buying it now, as he prepares to hand over the chief executive role, is more than a tidy final trade. It is a quiet statement about how he thinks durable value in artificial intelligence will actually look.
For ordinary investors, the point is not to “buy what Buffett buys”. It is to notice that even the most traditional value investor is happy to own artificial intelligence, as long as it lives inside a diversified, cash-rich platform that he can understand and justify. Markets will continue to argue about whether Alphabet’s price is too high, too low or somewhere in between. The better question is the one Buffett keeps asking after seventy years: which businesses can you hold in good times and bad because you understand how they make money and why they endure?
This material is marketing content and should not be regarded as investment advice. Trading financial instruments carries risks and historic performance is not a guarantee of future results.
The instrument(s) referenced in this content may be issued by a partner, from whom Saxo receives promotional fees, payment or retrocessions. While Saxo may receive compensation from these partnerships, all content is created with the aim of providing clients with valuable information and options.