Zscaler (ZS) CEO Jay Chaudhry told CNBC that he feels AI agents have supercharged cyberattacks, and at a pace that’s far quicker than most companies can respond. Enterprises have been sluggish in adapting to the AI agent threat, a bigger risk than the technology itself.

For the most part, generative AI has been mostly about large language models (like GPT-3.5 or GPT-4), but now the focus shifts to autonomous systems that can reason, decide, and act on their accord. 

These “agentic” models don’t wait around for prompts.

In fact, they have the ability to execute multi-step tasks, and now that robust capability is in play on both ends of the cybersecurity battle.

I’ve covered the tech space long enough to know that virtually every major innovation follows a similar pathway. 

At some point in a technology’s cycle, the downside risks become impossible to ignore.

Chaudhry argues that AI agents are entering this phase quietly and swifter than most bigwigs realize.

For attackers, AI agents continue to lower the skill barrier and scale. For defenders, they continue to compress response times while exposing major weaknesses in patchwork security systems.

The data throw weight behind his arguments.

In fact, a recent CrowdStrike survey found that 76% of organizations struggle to keep up with the speed and complexities of AI-led attacks. 

Moreover, 48% of leaders in the security space rank AI-powered attacks as their top ransomware threat.

For these reasons, Chaudhry feels customers see such developments taking shape in real time, even if Mr. Market hasn’t fully priced them in yet.

Zscaler CEO Jay Chaudhry warns AI agents will turbocharge cyberattacks, forcing enterprises toward security platforms.

Photo by Bloomberg on Getty Images

AI agents mark the shift from tools to autonomy

AI agents are less like chatbots and more like junior assistants that can work without constant supervision.

Traditionally, AI chatbots wait for instructions, answer questions, and then stop. On the flip side, an AI agent gets an objective and is then tasked with figuring out the next steps on its own.

More Tech Stocks:

• Morgan Stanley sets jaw-dropping Micron price target after event
• Nvidia’s China chip problem isn’t what most investors think
• Quantum Computing makes $110 million move nobody saw coming
• Morgan Stanley drops eye-popping Broadcom price target
• Apple analyst sets bold stock target for 2026

So an agent can plan, act, and check its work. It can also browse through the web, write code, and pull data every time something needs changing.  

That shift unlocks massive long-term productivity gains, which is why the tech punditry believes AI agents can become a massive market.

Tech experts’ predictions for AI agents market

• According to MarketsandMarkets, the AI agents market is forecasted to grow from  $7.84 billion in 2025 to $52.62 billion by 2030 (46.3% CAGR).
• According to Mordor Intelligence, the agentic AI market is set to grow from $6.96 billion in 2025 to $42.56 billion by 2030 (43.61% CAGR). 
• According to Grand View Research, AI agents could potentially jump to $182.97 billion by 2033, up substantially from $7.63 billion in 2025 (49.6% CAGR). 

AI agents are forcing a security reset

In the CNBC interview mentioned earlier, Chaudhry zeroed in on AI agents, particularly on their speed and scale, as well as the growing gap between attackers and defenders.

Chaudhry feels AI agents are catalyzing the “franchising” of cybercrime. 

Tasks that once required skilled hackers can now be easily automated and executed in seconds.

That worrying shift, he warns, constricts response times to the point where traditional approaches might simply break down.

That’s where Zscaler fits into the overall picture. 

With AI agents proliferating and threats intensifying, Chaudhry believes that cybersecurity has become more critical than ever.

He argues that only a unified, cloud-based platform can protect users, applications, and data in real time. 

That approach is mission-critical at a time when breach volumes are growing at a staggering pace. 

A recent Verizon Data Breach Investigations Report analyzed more than 22,000 incidents and 12,195 confirmed breaches in 2025, underscoring the razor-thin margins enterprises now have.

Major cybersecurity attacks in 2025

• A 16 billion-password “mega leak” turned out to be the largest credential exposure ever, essentially years of stolen logins compiled into a single cache, reported Gulf News. 
• The Salesforce-Drift supply chain breach involved hackers compromising a major third-party SaaS application by stealing authorization tokens, exposing nearly 1.5 billion CRM records, according to UpGuard. 
• A single major ransomware attack on UnitedHealth’s Change Healthcare disrupted U.S. health care systems, Reuters said, while exposing data linked to a whopping 192.7 million people.

Related: Rare Wall Street unanimous 2026 call means Nvidia is pressure point