Elon Musk, Big Tech get creative as they burn through AI money
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Allie Carl/Axios
The biggest tech companies are increasingly getting creative with their balance sheets as they finance their AI ambitions.
Why it matters: The AI race is looking strapped for cash.
Driving the news: SpaceX is merging with xAI.
- Oracle is issuing more debt and equity to fund its AI buildout.
- All the major tech companies that have reported earnings so far have also signaled plans to spend more money to keep up in the AI race.
What they're saying: AI companies are transitioning from "asset-light to infrastructure-heavy business models," Morgan Stanley analysts wrote in a recent note to clients.
- There's a risk that Oracle could need additional debt, Morgan Stanley notes, and its cash flow is expected to turn negative amid its AI data center construction efforts, according to Bloomberg.
- SpaceX and xAI are merging because the latter needs capital to compete more directly with the likes of Google and Meta in the AI race, Gil Luria, head of technology research at DA Davidson, told Axios.
- Microsoft's capital expenditure hit $37 billion for its most recent quarter, a 65% annual jump, leading to one of the biggest selloffs in the stock's history. (Meta's capex also increased, but all was forgiven by Wall Street because its AI-driven ad revenue also rose.)
Zoom out: Broadly, Big Tech companies are transitioning from low to high capex "with limited disclosure" about financial details, which could mean the expense of the AI buildout is being miscalculated, Morgan Stanley notes.
- Financial statements from the Big Tech companies don't always clearly show their increased spending habits, in part due to varied accounting practices for things like the depreciation timeline for chips and data centers.
- Michael Burry, of "Big Short" fame, called out depreciation as "one of the more common frauds of the modern era."
- According to his calculations, Oracle is overstating forward earnings by nearly 27% and Meta by nearly 21%.
Threat level: Investors are meant to value stocks based on their earnings, both current and future.
- That math breaks down when the underlying numbers can't be trusted in the first place.
The bottom line: AI capex estimates vary, but are expected to hit $5 trillion by 2030 according to McKinsey.
- For that math to work, the Big Tech companies have to find a way to make AI, or some other part of their businesses, profitable and fast.
