Why Meta's stock could see a 50% rally, thanks to an overlooked AI wild card

Meta’s AI spending has pressured the stock (down 6% over the past month), but Evercore ISI highlights an “overlooked AI wild card”: an industry compute shortage could let Meta monetize capacity for a premium. The analyst estimates a potential $22B annual gross revenue opportunity starting in 2027. If realized, this upside case could support an estimated 50% rally versus current levels.

Analysis

The market is still treating Meta’s AI buildout like a pure cost center, but the more important mechanism is that scarce compute can be monetized like industrial capacity. If management can credibly expose even a fraction of its internal infrastructure to third-party demand, Meta’s AI capex stops being just an expense drag and becomes an asset with salvage value — that can support a much higher terminal multiple even before cash earnings show up.

The biggest second-order winner is Meta’s own valuation framework: investors tend to discount heavy spend when the payback is vague, but they assign much more value once there is an explicit revenue-monetization path. That also creates pressure on peer hyperscalers to show similar optionality, while pure AI beneficiaries that rely on external cloud capacity may face tighter supply and higher pricing into 2026-27. The catch is that the incremental revenue may not be as high-margin as the headline suggests if it requires heavy power, depreciation, and support costs.

Near term, this is more of a sentiment catalyst than a hard fundamental inflection. The stock can re-rate over 1-3 months if management or channel checks validate that excess compute is real and sellable; over 6-18 months the bigger driver is whether capex intensity peaks while monetization becomes visible in guidance. The thesis is falsified if AI spend keeps rising without any externalization of capacity, or if industry supply normalizes faster than expected and collapses pricing power.

Contrarian angle: consensus is likely underestimating how valuable embedded infrastructure options can be, but it may also be overestimating the size of the revenue pool because gross compute demand does not automatically translate into durable profit. This is not a clean new business line until Meta proves utilization, customer demand, and accounting treatment. Until then, the trade is more about owning the optionality than underwriting the headline number.

Key Decisions for Investors

• Tactically accumulate META on post-rally pullbacks over the next 2-4 weeks; use the market’s obsession with AI spend as an entry point, with the thesis that optionality on compute can support multiple expansion before 2027 cash flow contribution is visible.
• Express the view with long-dated META calls or call spreads into 2027-2028 if liquidity is acceptable; risk/reward is best if the market continues to underprice the monetization option, but cap downside by avoiding outright leverage.
• Pair trade: long META / short a basket of AI spending beneficiaries that trade mostly on capex enthusiasm rather than near-term monetization (e.g., a small basket of high-multiple AI infrastructure names), looking for Meta’s valuation to close part of the gap once capacity monetization is acknowledged.
• Set an alert for the next two earnings cycles: if management discloses any external compute revenue, utilization metrics, or capex normalization, the re-rating window opens; if not, treat the thesis as a watch item rather than a conviction buy.
• If META breaks materially below recent support while capex guidance rises again without evidence of external demand, fade the idea and look for the market to keep penalizing free-cash-flow conversion rather than rewarding the optionality.