ZS Zscaler, Inc. : Bullish and Bearish Analyst Opinions

"The most disconnected of all the software [is] cybersecurity... nothing is replacing [it]... it increases the surface area and the models are going to have to be protected." The market wrongly sold off cyber stocks on the fear that AI writes its own code and doesn't need security. Ives argues the opposite: AI generates more code and threats ("surface area"), making CrowdStrike, Palo Alto, Check Point, and Zscaler more essential, not obsolete. LONG. Capitalize on the "misnomer" selloff. If AI agents become self-healing/self-securing faster than anticipated, legacy security seat-counts could compress.

"Software ETF (IGV) has been ripping... green every single day." Specifically mentions Palantir (PLTR) and CrowdStrike (CRWD) as "ripping." Despite macro fear, the market is bidding up software and cybersecurity. This suggests a "flight to quality" or a decoupling where tech is viewed as immune or essential (especially cybersecurity in modern warfare). Follow the momentum. The "fractal" correlation between BTC and Software suggests if one holds, the other runs. General market beta crash if SPY rolls over below 670.

A caller owns Zscaler along with CrowdStrike, Rubrik, and Cloudflare. ZS has issues regarding the Red Canary acquisition and stock-based compensation. While the stock is low (down from $250 to $156), the caller is over-exposed to the cybersecurity sector. Do not sell at the lows, but use any bounce to "lighten up" and reduce concentration risk. Sector-wide rotation out of cyber.

Zscaler stock is down ~36% YTD and fell further despite beating revenue estimates, driven by concerns over billings and AI disruption. The CEO argues that AI "Agents" will become the new users needing security (expanding TAM), but the market interprets AI as a threat to seat-based pricing (fewer human employees = fewer seats). Until ZS proves it can monetize "machine users" at the same rate as humans, the multiple will compress. WATCH. Wait for the "AI Agent" revenue thesis to show up in billings before catching the falling knife. If the CEO is right, the stock is massively oversold; if the market is right, churn will increase.

"I'm scratching my head and say where did market get it wrong? ... We have a Rule of 62... We know AI agents are coming for every employee. There'll be 50 to 100 agents... The best way to secure AI agents is Zero Trust technology." The market is selling ZS based on short-term ARR (Annual Recurring Revenue) concerns, ignoring the company's elite efficiency (Rule of 62). The CEO argues that the market is underpricing the next massive secular tailwind: the explosion of AI agents. Since these agents act like employees and need secure access to apps, the total addressable market for Zscaler's Zero Trust platform is set to multiply exponentially, making the current sell-off a misunderstanding of future demand. Long ZS as a contrarian play against the post-earnings sell-off, betting that the "Agentic Era" will drive a new wave of cybersecurity spending. Continued deceleration in ARR growth, market rotation out of high-multiple SaaS stocks, or competitors effectively capturing the AI agent security market.

Zscaler (ZS) is the author's top percentage loser, down 39%. The author interprets this massive drop not as a company-specific problem but as a sector-wide panic, presenting a deep discount buying opportunity. The author is buying more ZS, believing the sell-off is indiscriminate and that the company's prospects remain strong despite the significant price decline. A 39% drop could signal a severe degradation in the company's fundamentals or a permanent re-rating of its valuation multiples by the market. The author has not investigated the root cause. TICKER - DIRECTION

Over 80% of breaches are now "malware-free," meaning they exploit stolen credentials or other system weaknesses rather than code bugs. This trend makes active security solutions, like those provided by Zscaler, essential for enterprises, regardless of how secure their code is. The rise of AI is a tailwind, not a threat, to this business model. ZS is part of an "obvious secular bull market" in cybersecurity, and recent sector fears may present a buying opportunity. High valuation, intense competition within the cybersecurity sector, and the possibility that AI-driven code improvement does eventually reduce overall system vulnerabilities.

The author suggests the market is overly pessimistic about the impact of AI competition on Zscaler, implying the stock is undervalued heading into its earnings report.

Cybersecurity stocks (Zscaler, Cloudflare, CrowdStrike) dropped ~8% following the release of a new Anthropic tool with security features. The market perceives a threat that advanced AI models (like Anthropic's) could integrate security features natively, rendering standalone cybersecurity SaaS platforms less essential or forcing pricing pressure. SHORT (Disruption Risk/Sentiment Shift). Market overreaction to a single product announcement; these companies integrate AI to bolster their own moats.

DeepMind's CEO warned about the risks of "Agentic AI" (AI that takes autonomous action). Matt highlighted that as AI automates tasks, we lose control, exposing systems to massive cybersecurity risks. The deployment of Agentic AI necessitates a massive upgrade in security infrastructure to create "guardrails." This is a direct tailwind for cybersecurity firms like Palo Alto Networks (explicitly named) and peers. LONG. Cybersecurity is the necessary "second derivative" trade of the AI agent boom. Valuation compression if IT budgets tighten due to higher interest rates.

The speaker notes that despite bearish fundamentals from Brazil, a post by President Trump suggested "China is considering lifting the soybean count to 20 million tons for the season, an additional 8 million tons." This political intervention serves as a massive demand shock that overrides the previous bearish supply data (Brazil's record crop). The "flurry of buying" indicates the market is rapidly repricing the commodity based on this new diplomatic expectation. Long Soybean Futures (ZS) to ride the momentum of the US-China trade optimism. The demand increase is based on a social media post regarding "consideration," not a finalized trade deal; Brazil's supply remains structurally higher.

"Cybersecurity is now not 5 to 8%. You could argue from everything we're seeing, 20, 25% of now the potential budgets... I think is being mispriced, that AI is a headwind where in fact, I think it actually is going to change the view of the sector." Investors fear that AI tools (like Anthropic) will replace traditional software and security vendors. Ives argues the opposite: the deployment of AI use cases requires massive "end-to-end" security infrastructure, effectively tripling the addressable budget share for leaders like Palo Alto and CrowdStrike. He cites M&A activity (specifically mentioning a deal involving CyberArk and Palo Alto) as proof of this platform shift. LONG. Ives explicitly calls PANW a "double table pounder" and identifies CRWD and ZS as top picks that will benefit from this inflection point. The sector is currently "guilty until proven innocent" with negative sentiment; risk of earnings upsets (as seen previously with PANW); market skepticism regarding the timeline of AI monetization.

The Cybersecurity sector has been beaten down alongside software. Security is non-discretionary spend. These companies are approaching attractive buy zones. Solodin is watching them for entry but notes some (like ZS) have questions regarding multi-cloud competition. WATCH (Potential Longs). Further sector rotation out of high-multiple tech.

"There are massive security flaws in operating this entire system... spinning up this agent on your computer and then giving it access... Imagine if it posted all your passwords... I accidentally socially engineered my own human." The shift from "Chatbot" (cloud-hosted) to "Agent" (locally hosted with file access) obliterates the traditional security perimeter. If agents can read local files and trick users, legacy firewalls are useless. This necessitates Zero Trust architecture (ZS/PANW), Endpoint Detection (CRWD), and rigorous Identity Management (OKTA) to prevent authorized agents from performing unauthorized actions. LONG. As Agentic AI scales, the attack surface moves to the endpoint and identity layer. AI agents evolving faster than defensive software can patch vulnerabilities; "alert fatigue" for security teams.