A new survey from the State of AI Infrastructure report, covering over 1,400 senior IT leaders, reveals that 48% are prioritizing infrastructure with data residency and compliance controls, while 52% now use a hybrid cloud approach for AI.

These organizations face three major risks: jurisdictional, economic independence, and geopolitical.

The report argues that hybrid cloud (on-premises plus multicloud) lets enterprises balance public cloud power with sovereignty needs.

Google Distributed Cloud (GDC) is presented as a solution that brings Google Cloud into on-premises or edge environments.

GDC comes in two deployment models: air-gapped (fully disconnected, no remote shutdown possible) and connected (Google-managed lifecycle on existing hardware).

It offers a complete on-premises AI stack including optimized infrastructure, Gemini or open models, and cost-effective inference, enabling secure AI agents while keeping data under the customer’s control.

The key message is that organizations no longer must choose between data control and AI innovation.