Agentic AI blows up the attack surface as security moves into the infrastructure layer
The enterprise attack surface is expanding rapidly as agentic AI brings a constantly shifting cloud infrastructure into play. That pace is pushing security decisions down toward the virtualization layer, where policy can be enforced without slowing traffic. Perimeter defenses alone no longer hold, and enterprises that delay a broader lateral security program risk falling behind […] The post…
As agentic AI reshapes the enterprise attack surface, security measures are moving into the infrastructure layer. Umesh Mahajan, vice president and general manager of Broadcom Inc.'s Application Networking and Security Division, explained that perimeter defenses are no longer sufficient and that enterprises must deploy lateral security programs immediately.
Broadcom's solution involves an integrated software stack with shared context, delivered through vDefend and Avi Load Balancer products. Scale is a significant challenge, as AI workloads generate heavy east-west traffic and require enforcement at the hypervisor level to maintain high throughput and low latency. Broadcom is pushing security processing into the hypervisor to inspect traffic without adding latency, handling up to 75 terabits per vCenter cluster for firewalling and 17 terabits for IDS/IPS.
To protect these workloads, administrators need real-time visibility into authorized and unauthorized components, an aspect now integrated with private cloud modernization programs.
Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
