Cybersecurity Emerges as AI’s Next Investable Frontier

On September 14, CrowdStrike shares jumped 14% and Palo Alto Networks gained 13%, a sharp divergence from the broader technology selloff. Investors were not abandoning artificial intelligence. They were rotating toward the companies expected to contain its consequences.

The catalyst was a warning from Anthropic Chief Executive Dario Amodei that autonomous AI agents could become a serious security threat, combined with a more bullish prediction from Nvidia Chief Executive Jensen Huang. Speaking at Goldman Sachs’ Communacopia + Technology Conference on September 10, Huang called cybersecurity a likely “next major use case” for AI and pointed to continuous software testing, vulnerability discovery and automated defense as large new markets.

That thesis gives cybersecurity vendors a way to benefit even if the traditional AI trade cools. Chipmakers and cloud providers need customers to keep expanding data-center capacity. Security vendors can sell into the risks created by that expansion, including faster attacks, autonomous software with excessive privileges and sensitive corporate data flowing through unfamiliar models.

CrowdStrike is trying to turn that logic into a product moat. Its SafeMind system, introduced September 1, uses purpose-built models developed with Nvidia’s Nemotron and trained on Falcon security telemetry. The system pairs offensive agents that search for attack paths with defensive agents that test and close them, a loop that generic models may struggle to reproduce without comparable security data.

Palo Alto Networks is attacking a different choke point: identity. Following its CyberArk acquisition, the company is positioning its Idira platform to govern access for human users, machines and AI agents. Palo Alto says machine and AI identities now outnumber human identities 109 to 1, underscoring why permissions, rather than perimeter firewalls alone, are becoming central to enterprise defense.

The investment case is not risk-free. Cybersecurity companies still must prove that AI products lift recurring revenue rather than simply repackage existing automation, while lofty valuations leave little room for weak execution. But the mechanism is straightforward: every new autonomous system creates another identity, another access path and another attack surface.

That makes cybersecurity less a side bet on AI than an insurance premium on its expansion.