What changed
Nvidia Corp. introduced Scale-In, a new pillar in its AI networking infrastructure, aimed at securing, managing, and operating agentic AI factories. The platform provides secure access, high-performance data movement, tenant isolation, and more predictable performance as agents take on larger roles in AI systems.
What this means in practice
Scale-In is powered by Nvidia BlueField-4 and Nvidia DOCA, offering in-silicon security engines for out-of-band monitoring and telemetry collection from memory, storage, GPUs, and CPUs. It enables secure agent execution across enterprise systems by integrating with the OpenShell Reference Architecture, which includes Nvidia Sentry for continuous agent behavior monitoring and quarantine capabilities.
Limitations and availability
The supplied announcement does not specify exact deployment timelines or pricing details for Scale-In. It is part of Nvidia’s broader agentic AI security strategy, which includes the Open Agent Safety Platform and OpenShell Reference Architecture. Scale-In is designed for large-scale infrastructure needs and is compatible with existing systems using BlueField-4 and Spectrum-X Ethernet. The platform is intended for enterprise environments requiring robust security and operational control over agentic AI workloads.
