Why cybersecurity stocks will be net beneficiaries of the AI risks

Bank of America said AI-driven security risks should make cybersecurity stocks net beneficiaries, but it cut price targets across the group as software valuations compress. Zscaler's target was cut to $175 from $335, SailPoint to $16 from $27.50, and SentinelOne to $16 from $18, reflecting slower billings growth, longer sales cycles, and tougher competition. The bank also flagged AI inferencing demand as a growth tailwind for Akamai and Fastly.

Analysis

The key market implication is not that cybersecurity is broadly attractive, but that the industry is splitting into two lanes: vendors with embedded workflows and recurring cash generation can keep compounding, while point solutions with long sales cycles are becoming financing-risk assets. In this tape, valuation compression matters more than AI-linked demand uplift, so the market will reward names that can convert AI-driven demand into near-term FCF rather than just pipeline.

The second-order winner is likely to be the security layer that sits closest to traffic and infrastructure, not the identity or endpoint vendors that require deeper organizational change to displace. If larger platforms continue bundling security into cloud and productivity suites, standalone vendors face margin pressure from discounting and slower deal closure, especially over the next 2-3 quarters. That argues for a relative-value approach: favor infrastructure-adjacent and usage-based names over subscription-heavy pure plays.

Akamai and Fastly may benefit more from AI inference traffic than the market is currently pricing, but the cleanest expression is not a directional bet on “AI edge” as a concept. The real catalyst is whether inference growth shows up in utilization and pricing before capex expectations reset lower; if so, multiples can re-rate over the next 6-12 months. By contrast, names with weaker balance sheets or dependence on enterprise budget expansion could underperform even if sector fundamentals remain okay.

The contrarian read is that this may be a quality filter disguised as an AI theme. Investors are likely overpaying attention to AI-driven demand and underpaying attention to procurement friction, platform bundling, and the fact that AI security spend often gets pulled from existing IT budgets rather than added as incremental spend. That makes the upside for the strongest cash generators more durable than the upside for growth-at-any-price software.

Key Decisions for Investors

• Long AKAM vs short ZS for 3-6 months: favor edge/inference exposure over a name facing multiple compression and slower billings; target relative outperformance if enterprise budgets stay tight.
• Avoid initiating new longs in SAIL and S on strength until evidence of FCF inflection: these are more vulnerable to duration compression and longer sales cycles over the next 2 quarters.
• Buy AKAM on 5-8% pullbacks with a 6-12 month horizon: best risk/reward if AI inference traffic translates into measurable utilization and pricing power; cut if growth fails to reaccelerate by the next earnings cycle.
• Pair long AKAM / short software ETF or broad SaaS basket: isolates the AI-infrastructure beneficiary against the broader valuation reset in software, with less single-name execution risk than a pure short on ZS.
• Use call spreads rather than stock in FSLY: if inference demand is real, upside can be sharp but capital structure and execution risk argue for defined-risk exposure only.