The new math defining AI winners and losers
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Sarah Grillo/Axios
Microsoft had one of its worst selloffs in history after it increased its AI spending plans. Meta did the same, but it was rewarded by investors.
Why it matters: The wildly different market reactions to Meta and Microsoft earnings reveal the new calculus investors are using to evaluate AI winners and losers.
State of play: Microsoft reported solid earnings, with revenue and profit beating expectations, but its stock fell nearly 12%.
- Cloud revenue was up 39% from a year ago, but that was driven by non-AI workloads. Capital expenditure hit $37 billion for the quarter, a 65% jump.
- That mismatch is the issue: If a company is spending a lot, fine, but it needs to be spending on AI investments that will make it money now.
- Meta beat on earnings, and its stock jumped even as it outlined over $115 billion in planned AI spend for the year, because investors believe its AI investments will fuel more growth in its biggest revenue driver: ads.
What they're saying: Meta also has a leg up because it is not a software company. Investor sentiment on software is "negative" to the point of the sector being "extremely radioactive," Anurag Rana, senior analyst with Bloomberg Intelligence, tells Axios.
- Wall Street has software in the doghouse over concerns that with the explosion of Claude Code, software will be obsolete.
- The math behind what it takes to value a company completely breaks in a world where AI can replace software entirely.
Zoom in: At the heart of the reaction to both stocks is a simple framework.
- Analysts use discounted cash flow models to value companies.
- These models rely heavily on future cash flows and a "terminal value" which is how much the business is worth beyond the forecast period.
- For big tech companies investing big money into AI, that terminal number can look great, because they stand to benefit from their investments in the technology longer term.
- This is why Meta was rewarded this earnings season: Its core business, advertising, is generating strong predictable cash.
- Analysts think AI will eventually add, not erode, Meta's future cash flows, making its valuation math easier to justify even with its huge spending.
Threat level: For software, the problem is some investors view AI as a complete replacement, which would demolish cash flows.
- This is why some investors see the heavy AI spending and uncertain return timeline from Microsoft as a valuation risk.
Thought bubble: It's interesting investors are rewarding Meta for a business that will either be helped by AI, or not hurt if AI proves to unprofitable.
- Feels like part of a broader theme also seen in Alphabet's outperformance: Investors like companies that are either making money from AI today, or are fine even if they never do.
