Akamai Brings Security Inside AI Factories with NVIDIA - GlobeNewswire

Akamai and NVIDIA have expanded their security collaboration to integrate advanced infrastructure-level protection into AI factories, addressing the growing cybersecurity risks associated with AI work... Read original →

The Akamai-NVIDIA integration of Zero Trust into AI factories marks a paradigm shift from perimeter-based to workload-native security in accelerated computing. Technically, merging Guardicore’s micro-segmentation with BlueField-4 STX’s DPU offloads enforcement from CPUs/GPUs, rendering legacy virtual firewalls obsolete and forcing upstream chipmakers to embed security primitives. This preempts stringent EU AI Act and NIST AI RMF mandates on runtime integrity, cutting operational risk for global enterprises. Competitors like Palo Alto Networks will likely accelerate DPU-focused M&A or deepen AMD/Xilinx alliances. Within 18 months, ‘security-as-infrastructure’ will become table stakes for AI clusters—data centers in Taiwan, China and Hong Kong, China lacking such capabilities risk exclusion from premium AI foundry ecosystems.