Last week, Google’s parent company, Alphabet, lost approximately $500 billion in market value in a single day. Investors were rattled by slowing advertising growth and the company’s enormous spending on artificial intelligence.
But buried in Alphabet’s quarterly financial disclosures was something arguably more interesting.
More than one-third of the quarterly increase in Google Cloud’s backlog of future contracted revenue came from a company that Google itself had helped create.
You read that correctly.
Meet Crux AI
Earlier this year, Alphabet partnered with investment giant Blackstone to establish Crux AI, a company designed to provide AI computing infrastructure.
Alphabet owns a minority stake in the venture and has agreed to provide up to $10 billion in financial backstops to support its data center and energy infrastructure agreements.
Crux AI, in turn, has signed a multiyear contract with Google worth approximately $16.8 billion.
That contract now appears in Google’s cloud backlog, the commitments from customers that the company expects to turn into future revenue.
Between June and September, Google’s cloud backlog increased from $513.9 billion to $560.8 billion. The Crux AI contract accounted for roughly 36% of that increase.
Think about that for a moment.
Google helps create an AI company, provides financial support for its infrastructure, and then counts that company’s business among the evidence of growing demand for Google Cloud.
The transaction isn’t necessarily improper. Crux AI may provide valuable services to paying customers. And Google has disclosed the relationship in its financial filings.
But it does raise an uncomfortable question.
How much of the AI industry’s extraordinary growth reflects demand from independent customers, and how much is being fueled by the industry’s own investments?
Money Going in Circles
This isn’t an isolated concern.
Across the AI industry, technology companies are investing in startups that subsequently spend enormous sums purchasing computing resources from those same companies or their partners.
The arrangements can make economic sense. AI startups need computing power, and infrastructure providers want to secure customers for the data centers they’re spending billions to build.
But they also create complicated financial dependencies.
Consider what happens if demand for AI services fails to materialize at the levels everyone expects.
The startups may struggle to pay their bills. The cloud providers could lose anticipated revenue. And the companies financing the infrastructure may discover they’re exposed to losses on both sides of the transactions.
In Alphabet’s case, the company has disclosed more than $1 trillion in future purchase, lease, and investment commitments, alongside approximately $98.5 billion in contingent financial obligations.
That’s a staggering amount of money riding on continued growth.
Is AI Becoming a Self-Fulfilling Prophecy?
None of this means AI is a scam. The technology is real, its capabilities are remarkable, and companies are finding genuine uses for it.
But technological breakthroughs and profitable investments aren’t the same thing.
The internet transformed the world, yet that didn’t prevent the dot-com bubble from destroying enormous amounts of shareholder wealth. Many companies were right about the technology’s future and spectacularly wrong about the economics.
AI could follow a similar path. The question is whether the demand will ultimately justify the extraordinary amounts of money being invested in it.
And when a company can help finance its own customers, then point to those customers as evidence of booming demand, perhaps investors ought to look a little more closely at where all that money is coming from.
Jayson L. Adams is a technology entrepreneur, artist, and the award-winning author of The Quantum Mirror, Ares, and Infernum.
His novels blend high-stakes science fiction, psychological tension, and character-driven suspense. Explore the books at www.jaysonadams.com.