The market's AI contradictions
Speakers
Deirdre Bosa — Anchor/Reporter, CNBC Tech Check
— CNBC anchor, tech reporter
Summary
- The market is currently grappling with a massive contradiction in the AI trade: Software stocks are crashing because investors fear AI will replace them, yet Big Tech (Hyperscalers) are also being punished for spending hundreds of billions to build that very AI.
- Four large hyperscalers have announced over $600 billion in capital expenditures (Capex) for this year alone.
- The market is in a "wait and see" mode, looking toward potential IPOs from OpenAI (rumored $1 Trillion valuation) and Anthropic to provide hard data on margins and burn rates to validate the AI business model.
Trade Ideas
Despite strong top-line growth, these stocks are being punished. Amazon is down 9% despite AWS growing at its fastest pace in three years. Investors are spooked by the massive capital expenditure required to compete in AI. While revenue is growing, spending is growing faster, compressing margins and free cash flow. The market is struggling to reconcile the cost of building AI with the future payoff. Meta is spending $0.50 of every revenue dollar on infrastructure. Meta's operating margins fell from 48% to 41% year-over-year, and free cash flow dropped 15%. Amazon has a backlog up 38% but is still seeing its stock drop. If these companies prove that the massive spend leads to immediate monetization, sentiment could reverse quickly.
OpenAI is preparing for a potential $1 Trillion IPO, and Anthropic is reportedly in early talks to go public. Currently, the market is guessing about the profitability of pure-play AI. When these companies file for IPOs, they will be forced to open their books. This will reveal the "hard numbers"—revenue, margins, and burn rates—finally confirming or denying if the AI business model makes economic sense. Reports of upcoming public filings. If the disclosed numbers show unsustainable burn rates or low margins, it could crash the broader AI sentiment across the entire market.
