CrowdStrike CEO and founder George Kurtz is warning that a potential slowdown in frontier AI development would not eliminate cyber threats already operating at machine speed. His message comes as governments, AI companies and researchers debate whether the development of increasingly capable models should be deliberately slowed, while cyber attackers are already using AI to accelerate reconnaissance, exploitation and intrusion.

Cyberattacks Are Already Moving at AI Speed

Kurtz argues that the central cybersecurity challenge is no longer simply how powerful the next AI model will become, but how quickly autonomous systems can execute attacks using capabilities that already exist. His warning is that “pacing what comes next doesn’t secure what’s already here,” highlighting the gap between regulating future AI capabilities and defending against threats currently active in enterprise environments.

CrowdStrike’s latest threat research illustrates the acceleration. AI-enabled adversary operations increased 89% year over year in 2025, while the average eCrime breakout time fell to 29 minutes. The fastest observed breakout took only 27 seconds, underscoring how quickly attackers can move once they gain an initial foothold.

Why Slowing Frontier AI Is Not a Complete Security Solution

Calls for a slower pace of frontier AI development have intensified following recent incidents involving autonomous AI agents and concerns that advanced models can increasingly perform complex cyber tasks. However, even a coordinated slowdown would not remove existing models, deployed agents, automated attack infrastructure or the accumulated techniques already available to malicious actors.

That distinction is particularly important for enterprises. Traditional cybersecurity defenses often rely on human analysts investigating alerts and responding to suspicious activity. When an autonomous attacker can reason, plan and execute an intrusion faster than a human can review an alert, the defensive model has to change as well.

CrowdStrike Positions AI Security as a Growth Opportunity

The shift is also becoming strategically important for CrowdStrike. The company reported $1.47 billion of revenue in the second quarter of fiscal 2027, up 26% year over year, while ending annual recurring revenue reached $5.84 billion. Net new ARR reached a record $333 million, increasing 51% from a year earlier.

CrowdStrike is responding by expanding its agentic security capabilities and investing in AI systems designed to detect, investigate and respond to threats autonomously. The company has also been developing cybersecurity-focused frontier models with NVIDIA, reflecting a broader shift toward using AI not only to identify threats but also to automate remediation.

The next phase of the cybersecurity market will increasingly be defined by the speed of the contest between autonomous attackers and autonomous defenses. For investors, the key indicators will be whether CrowdStrike can convert accelerating AI-related threats into sustained platform adoption while maintaining detection accuracy, response speed and financial growth. The broader policy debate over slowing frontier AI may continue, but the cybersecurity race is already underway.