Investors Are Underestimating This Incredibly Cheap Artificial Intelligence (AI) Stock. Buy It Before It Joins the $2 Trillion Club

Meta shares jumped nearly 9% on July 1 after reports (via Bloomberg) that the company may sell excess AI cloud computing capacity and offer access to its AI models for customers renting compute. The bullish case is supported by strong operating momentum: Q1 revenue rose 33% YoY to $56.3B, alongside AI-driven improvements (e.g., Meta AI user sessions up double digits and client issue resolution 20% faster). Despite the positive AI monetization narrative, heavy capex remains a risk—Meta is on track for $135B capex in 2026 (vs $72.2B last year)—but the article argues faster earnings growth could drive meaningful upside over the next three years.

Analysis

The market is likely keying off utilization optics, not a near-term revenue step-up. If Meta can monetize spare compute, the economic win is leverage on already-committed capex: higher depreciation absorption and a better ROIC story without requiring a fresh investment cycle. That matters more for the multiple than for this year’s P&L, because investors have been penalizing the stock for capex intensity rather than absolute growth.

Second-order, this strengthens Meta’s relative position in performance advertising: better internal AI economics can subsidize product improvements and widen the gap versus peers that still have to buy comparable inference capacity from third-party clouds. The spillover to the AI stack is mixed—higher utilization at a mega-buyer supports confidence in the GPU demand cycle, but it also suggests AI capacity is becoming less scarce, which can eventually pressure pricing power for cloud/infra providers.

The contrarian risk is that the addressable opportunity is overstated. Meta does not have a natural enterprise distribution advantage, so external AI-cloud demand may start as a niche developer/adjacent workload business rather than a durable third pillar. Over the next 1-3 months, the key falsifier is any capex raise without a matching improvement in operating margin or cash conversion; over 6-18 months, the thesis breaks if ad growth normalizes while AI monetization remains mostly internal rather than a real revenue line.

Key Decisions for Investors

• Accumulate META only on pullbacks over the next 2-4 weeks; prefer a 3-6 month call spread ~10-15% out-of-the-money to express upside from a re-rate while capping capex headline risk.
• Pair trade: long META / short GOOGL for 1-3 months as a relative AI-ad execution trade; thesis is Meta’s AI tools monetize faster, while Alphabet bears more of the cloud/ads execution burden.
• Do not chase the initial pop if management does not confirm external AI-cloud bookings on the next earnings call; set a hard invalidation if capex is raised again without a corresponding margin inflection.
• Watch NVDA as a secondary beneficiary only if Meta’s disclosures show sustained GPU utilization and no slowdown in inference spend; otherwise treat the signal as sentiment-driven, not a new demand leg.