Zscaler Blew Up in May. Its AI Security Business Is the Real Story.

Let's be real: if you looked at Zscaler's chart in late May, you probably threw your phone across the room. Down 31% in a single day. A guidance reset that lopped more than a third off the market cap. From a 52-week high of $337 to where it sits today at $141 — that's a 58% round trip that'd make even the chillest portfolio manager reach for the Tums. And yeah, the headline numbers look rough. Revenue still growing at 25% isn't exactly a crisis, but the market heard "FY2027 ARR growth decelerating to 16-17%" and decided that was the only number that mattered. But here's the thing about panics — they tend to throw the baby out with the bathwater. And in Zscaler's case, that baby is an AI security business that's quietly become one of the most explosive growth stories in enterprise software.

Let me break it down for you. Zscaler's AI security revenue just crossed $400 million in annualized run rate. That's not a nice-to-have side project — that's a full-on business line growing at over 80% year-over-year. Here's the kicker: they hit that milestone three quarters ahead of their own internal forecast. In a market where every cloud security company is screaming about AI, Zscaler actually delivered. This isn't vaporware. This is real revenue, real customers, real contracts. The company's been quietly building what they call the "AI Security Cloud" — a dedicated layer inside their zero trust platform that protects how enterprises actually use AI tools in production. Not just blocking ChatGPT at the firewall, but governing data flows to and from dozens of AI services, detecting model manipulation, and preventing sensitive data from leaking through prompt injection attacks. It's the kind of security that didn't exist two years ago and is now table stakes for any Fortune 500 running AI workloads.

And the market is starting to wake up to it. At Zenith Live 2026, Zscaler's annual customer conference, the company dropped a series of announcements that should have moved the stock more than they did. AI Broker — a policy engine that sits between your enterprise data and every AI model your employees are using, whether it's OpenAI, Anthropic, Google Cloud's Vertex AI, or open-source models running on Databricks. Endpoint AI Security — which extends zero trust down to the device level for AI-powered applications. And Project AI-Guardian, a partnership with AWS, Google Cloud, OpenAI, and Databricks that's essentially building the industry standard for securing AI pipelines. That's not just product releases. That's Zscaler planting a flag and saying the entire AI security market runs through their cloud.

Now, let's talk about valuation, because this is where the math gets interesting. Zscaler's trading at 7.2x sales — the cheapest it's been since 2023. Its forward P/E is 31, which for a company growing revenue at 25% with 77% gross margins and nearly a billion in free cash flow over the last twelve months is not exactly nosebleed territory. They've got $3.5 billion in cash against $1.9 billion in debt — net cash positive in a high-growth sector where most competitors are still bleeding red ink. The operating margin is negative, sure, but we're talking about a company investing aggressively into R&D and go-to-market at exactly the moment when its biggest growth vector — AI security — is accelerating. Wall Street analysts aren't stupid. Forty-four of them cover this stock, and the consensus is a Buy with a mean price target of $192.58. That's 36% upside from here. The most bearish analyst on the street still sees $150 — basically where it's trading now.

Look, I'm not going to tell you there's no risk. The guidance reset was real. The macro environment for enterprise tech spend is uncertain. If the Fed stays hawkish and CIOs tighten budgets, high-growth names like Zscaler get compressed further. And yes, the deceleration from 25% to 16-17% ARR growth is a legitimate concern — at some point, the AI security tailwind has to carry enough weight that the core zero trust business doesn't have to do all the heavy lifting. But here's what keeps me coming back: Zscaler is the infrastructure layer for one of the biggest secular shifts in enterprise IT since the move to the cloud. Companies are deploying AI into their workflows faster than they can secure it, and Zscaler built the pipes. The stock got punished for growing less fast than investors expected. But the underlying business — the one that went from zero to $400 million in AI security ARR in about 18 months — is stronger than ever. Sometimes the best trade is buying what everyone else just sold.

Disclosure: The Signal holds no position in ZS. Positions may change. This is not financial advice.