I pride myself on being right about the future, especially when it comes to money and technology.
But I was dead wrong about Alphabet $GOOG.
Last year I was certain Google was dead in the age of AI. I put out a sell at InvestorPlace in May. I repeated it here in October, comparing Google to AT&T.
I saw Google as a lazy, bottom-line oriented company. Its search dominance would wane in the age of AI. It was stuck at a 10% share of the cloud market, despite enormous investment. Under CFO Ruth Porat, Google eliminated most of its “other bets” and became as nasty a place to work as any in Silicon Valley.
I even disparaged Google’s strengths, hence the AT&T comparison. Its entrance into media, through YouTubeTV, looked cheap and second rate. Google Fiber had stalled out, the company never offering service at my house after years of promises. Waymo looked like a joke.
What happened?
The answer lies in this chart from Synergy Research, my go-to source on all things cloud. Google has been steadily gaining share in the cloud market, at the expense of Amazon.Com $AMZN and smaller players.
The data center market has been transformed, from the clean green earning machine it was in the 2010s, when Google could readily recycle its water and energy, into the second greatest threat to the planet after oil, which is how it’s seen today. The fact that it’s a money sink plays to Google’s advantage, as its enormous cash flow ($151 billion) and financial strength ($25 billion in debt on $450 billion in assets) can handle the load.
Then There’s Gemini
A bigger reason why the stock is up an amazing 67% this year, taking its market cap to $3.84 trillion, and making co-founder Larry Page the second-richest name on the Forbes 400, is Gemini.
Axios has a cute illustration on this, a Google “G” following an OpenAI logo like a lion after a gazelle. But the reviews on Gemini 3 are insanely great. Wei-Lin Chiang, who co-founded LMArena, a crowdsourced AI evaluation platform, told The Verge Gemini has a “clear lead” in coding, writing, and visual comprehension.
Gary Marcus, my favorite AI analyst, had the best reaction of all. Commenting on the Administration’s flirting with a bailout of the AI industry, he said not to worry. “If OpenAI fails, Google can readily step in, no bailout required,” he wrote. The Street recently claimed OpenAI is worth $500 billion despite its enormous losses, but a $4 trillion company could take that on. Trouble is any OpenAI transaction, at any price, will instantly pop the AI bubble and leave everyone poorer. But I’m sure Larry can eat chicken for a few weeks, and Google will be left controlling the field.
What Does This Mean?
The idea of an “AI bubble” is based on the idea that there won’t be enough work for data centers to keep them all operating. They will become like the “dark fiber” lines near my house in the early 2000s. Buddy can you spare a gigabit?
But what mainly happens both during and after a bust is consolidation. Mergers that wouldn’t have been imagined become done deals faster than you can push the paperwork. That’s what happened to the banks in 2008.
If the Administration is right about anything, it’s that data centers are the banks of this decade. Once the bubble pops they’ll go down faster than COVID patients in March 2020. Google’s value may fall back to $1 trillion, down nearly 75%. But it will still be Warren Buffett in the bust (or Mr. Potter if you prefer a Christmas analogy), picking up cloud market share for a song.
I think buying Alphabet shares here is foolish. Sorry I done y’all wrong. But they’ll get through, and for the AI economy that’s all that matters.
