The agentic AI era is forcing a fundamental redesign of enterprise infrastructure, extending to on-premises data centers. Regulated industries and governments require AI-ready infrastructure capable of running advanced AI models locally, without compromising data sovereignty. This demand challenges the cloud-native assumptions of the past decade.
Muninder Sambi, Google's VP and GM for Networking and Security, stated that enterprises previously faced a choice between data sovereignty and advanced cloud capabilities. "With Google Distributed Cloud, we are actually bringing the power and the intelligence of Gemini and all that Google has to offer for an on-premises environment," Sambi explained.
Google is partnering with Nvidia and Dell to bring Gemini foundation models and Gemini Flash models to air-gapped and connected on-premises environments. This enables enterprises to deploy sovereign AI workloads without data leaving their premises, utilizing hardware accelerators that Sambi likens to an "AI engine."
Kubernetes is emerging as the central control plane for AI, supporting workloads from training to inference. Drew Bradstock, Google's Senior Product Director for Kubernetes, noted that the open-source community's adaptation of Kubernetes for AI compatibility is now crucial for hybrid environments.
Furthermore, Google is adapting its tooling and interfaces to prioritize AI agent consumption, recognizing that AI agents are increasingly handling engineering tasks. "The north star for user experience for our group isn’t people anymore - it’s actually how we give the best experience for agents," Bradstock said.