This Underrated Artificial Intelligence (AI) Infrastructure Stock Has Surged 80% in a Year. It Can Still Surge 53%.

Jabil reported fiscal Q2 revenue of $8.3B (+23% YoY) and adjusted EPS of $2.69 (+39% YoY), both beating expectations. Management raised fiscal 2026 EPS guidance to $12.25 from $11.55 and now forecasts AI revenue to grow 46% to $13.1B (vs prior 35%); capacity utilization is 75% with a target of 80% and adjusted operating margin is expected to exceed 6% (vs 5.7% prior). The company is in talks with a third hyperscaler that could become a major fiscal 2027 contributor, and Jabil has invested $500M to expand cloud/AI manufacturing capacity.

Analysis

Jabil sits at the intersection of two durable trends: hyperscaler consolidation of specialized AI infrastructure and a re-shoring/outsourcing wave for higher-precision electromechanical assembly. The key non-obvious leverage is margin optionality from service-augmented manufacturing (engineering-to-order racks, thermal validation, logistics for blade rollouts) — these functions convert capital-intensive revenue into stickier, higher-margin annuities if executed well.

Competitive dynamics will increasingly be decided by execution on thermal IP and supply of scarce subsystem components (cold plates, high‑capacity power distribution, vibration-tolerant connectors) rather than conventional EMS scale alone; incumbent server OEMs and niche thermal specialists are the realistic short-term gatekeepers. Over 12–36 months the bigger threat is customer vertical integration: a hyperscaler choosing to internalize rack assembly or award exclusive long-term contracts to a single supplier could flip Jabil’s current optionality into concentration risk.

Catalysts to watch are order wins/losses from hyperscalers, quarter-to-quarter capacity utilization inflections, and component lead-time normalization; each can compress or expand realized margins materially in a single cycle. Tail risks include a GPU spending pause that cascades into server order cancellations and technological shifts (e.g., immersion cooling standards or open-hardware form factors) that render current factory tooling suboptimal within 18–36 months.

Key Decisions for Investors

• Long JBL equity (6–18 months): initiate a core position sized for event risk (10–15% portfolio active weight cap). Add on confirmed hyperscaler contract close and sequential utilization >80%; use a 20% trailing stop to protect against single-customer reversal.
• Call-spread on JBL (9–12 months): buy a 12-month call and sell a higher strike to fund cost — asymmetric upside if AI server FCF sustains and margins re-rate; cap premium loss to the net debit (target 3:1 reward-to-risk if guidance confirms).
• Pair trade — long JBL / short FLEX (6–12 months): structural premium for higher-value AI-specific assembly favors JBL while generalist EMS faces margin compression. Size the short to keep net delta small; unwind on divergence >30% vs entry.
• Event hedge around hyperscaler decision (weeks): buy JBL puts or long risk-reversal (long put + short call) sized to offset 50–75% of equity exposure if the rumored hyperscaler deal fails to close within announced window.