Prediction: This Cybersecurity Stock Will Double After the AI Disruption Fears Fade

Zscaler reported fiscal Q2 FY2026 revenue up 26% YoY to $816M and ARR up 25% YoY to $3.4B; consumption-based offerings comprised >25% of new business and related ARR grew >100%. The company is marketing an AI Protect suite to address AI-driven threats (prompt injection, data leakage, shadow AI) and inspects >500 billion transactions daily across 160 data centers. Shares trade at ~7.3x sales versus a five-year average of 17.9x, implying a potential re-rating if AI adoption drives demand and continued revenue/ARR growth. Approximately 4,400 large enterprises are customers, indicating meaningful expansion runway.

Analysis

The real earnings lever to watch is consumption elasticity, not headline bookings: AI agents turn security from a seat-license sale into a high-frequency telemetry and policy-enforcement consumption stream. Vendors with in-line inspection fabrics and low-latency peering will be able to monetize per-interaction flows, creating a durable ARPU uplift that compounds faster than legacy license rollouts.

Second-order winners include cloud infra and networking suppliers that capture incremental egress, accelerated peering, and managed-inspection fees; conversely, appliance-centric vendors will face margin pressure as customers trade capital spend for operational, inline protection. Hardware for on-prem inference (NPUs, NIC offload) will see bifurcated demand — premium silicon for secure inference at the edge, and commoditized volumes for bulk training fueling GPU/cloud demand.

Key risks are structural rather than cyclical: hyperscaler standardization of model-security APIs or rapid enterprise insourcing could compress vendor TAM and margins, and regulatory mandates (data residency, model auditability) could reorder procurement priorities within 6–18 months. Near-term catalysts to watch are multi-cloud security partnerships, material enterprise consumption disclosures, and precedent-setting breach events tied to agentic AI; any of these could re-rate multiples quickly.

Positioning should therefore be asymmetric: favor firms that can both grow per-transaction revenue and keep gross-margin per byte resilient. Execution cadence (sales motion adapting to consumption billing, customer success to prevent churn) is the single biggest operational risk; miss that and even a large TAM won’t translate into expanded public multiples within a 12–24 month window.

Key Decisions for Investors

• Long ZS equity (or 12–18 month call spread): Initiate on a material pullback or on announcement of a large hyperscaler partnership. Position size 3–5% of active risk with a stop at -25% and a target of +40–80% over 12–24 months; use a defined-risk call spread if you want to cap downside (buy 12–18 month ITM call, sell higher strike to fund premium).
• Pair trade — long ZS / short PANW (or a legacy appliance-rated peer): Size 1:0.6 to reflect revenue base differences, horizon 9–18 months. Expect asymmetric upside if consumption re-accelerates; set combined stop if pair moves against you by 18% to limit model/sector beta risk.
• Long NVDA LEAPS (9–18 months) as a convex hedge to AI compute demand: buy calls to capture upside in cloud-side inference and training capex. Risk: high IV — allocate 2–3% of portfolio with max loss = premium; target 2x–4x return if AI infra continues to scale.
• Tactical event hedge: buy 3–6 month protective puts on ZS sized to cover >50% of position cost ahead of major earnings or product partnership announcements; cost-to-hedge should be weighed against expected alpha from the catalyst (acceptable if <3% of position value).