Security teams are increasingly turning to virtual patching, a technique that shields applications from known vulnerabilities at the network layer while the official software fix is developed and deployed, according to executives at F5 Inc. The shift comes as automated scanners are able to weaponize newly disclosed flaws within hours, often before a vendor can release a patch. This collapse in the window between disclosure and exploitation has made the traditional "Patch Tuesday" head start largely obsolete.

John Maddison, chief marketing officer and head of technology alliances at F5, said the company's customers are demanding this type of protection more than ever. F5, an application delivery and security vendor that turns 30 this year, sees AI-powered security as the necessary response to close the gap. Maddison stated that protections residing in content delivery networks or web application firewalls must be AI-powered going forward, capable of inspecting traffic and producing real-time defenses against threats such as zero-day exploits.

Maddison made these remarks during an interview at the Fal.Con event, in a broadcast from theCUBE, SiliconANGLE Media's livestreaming studio. He discussed how F5's partnership with CrowdStrike treats network appliances as endpoints. The companies have embedded CrowdStrike's Falcon sensor directly onto F5's BIG-IP appliances, which sit at the network perimeter. This integration allows CrowdStrike to monitor those appliances the same way it would a laptop or a cloud workload.

Before the partnership was formally announced, more than 200 customers were already running the integrated setup. F5 measured the performance overhead of the security integration at just 1% to 2%, a trade-off Maddison said customers are willing to accept. He noted that this addresses the traditional conflict between high-performance networking and thorough security inspection.

The urgency behind these measures is underscored by threat research from CrowdStrike, which found that attacks from AI-enabled adversaries rose by 89% year over year. The research also showed that the average time for an eCrime breakout has fallen to 29 minutes, with the fastest recorded breakout occurring in just 27 seconds. These figures highlight the speed at which modern attackers operate.

F5 is also applying its security approach to the emerging challenge of AI gateways. This week, the company announced a partnership with Salesforce Inc.'s MuleSoft to address this area. Maddison explained that the focus is less on the underlying information itself, which organizations already possess, and more on managing what is done with that information and how to remediate issues. The full video interview was part of theCUBE's coverage of the Fal.Con event.

More cybersecurity news from TechManNews.