Cisco takes Splunk AI behind the firewall with Nvidia-powered POD

Cisco's move extends Splunk beyond the monitoring role enterprises already know, into the environments where they are most reluctant to hand data to a public AI service. The AI POD — Cisco infrastructure paired with Nvidia accelerated computing and Kubernetes software — is aimed squarely at private-cloud, on-premises and air-gapped setups, which is where regulated or otherwise sensitive machine data tends to live. Splunk AI Assistant is already available; Agent Launchpad is expected later this year.

The second piece is governance. As companies move from small AI experiments toward fleets of autonomous agents, the cost of those agents becomes harder to pin down. Cisco's Tokenomics capabilities are designed to track agent spending, attribute usage and forecast consumption, which gives Cisco a second angle on the enterprise AI budget alongside the infrastructure itself.

The open question is whether that budget materializes on Cisco's timeline. The company has yet to disclose pricing or committed deployment volumes, and the shares already trade at a premium: $109.73 versus a GF Value estimate of $74.62, or 47.05% above that estimate. The products broaden Cisco's addressable market, but the valuation is likely to hinge on adoption and revenue showing up before investors are asked to keep paying for it.