NVIDIA Confidential Computing expands, enabling frontier model builders to securely scale across global infrastructure
NVIDIA adds a Confidential Bare Metal option to its Confidential Computing platform so advanced AI developers can run sensitive workloads on dedicated third-party hardware while keeping control of models and data. It spans three deployment types, Grace Blackwell NVL72, Rubin and DSX OS, with support from Google Cloud and Microsoft Azure.
What the article says
In a post dated 21 October 2026, tied to the GTC Berlin keynote, NVIDIA presents the expansion of NVIDIA Confidential Computing, the layer that protects models and data while they run. It is part of the effort to secure AI agents and accompanies the OpenShell Reference Architecture announced the same day.
The central addition is Confidential Bare Metal, which lets model builders run sensitive workloads on dedicated third-party infrastructure while keeping control of their models and data. The unified architecture has three modes: bare metal, Confidential Containers for Kubernetes and Confidential Virtual Machines. It extends to Grace Blackwell and Rubin systems. Cohere plans to integrate it into its Model Vault, and Google Cloud extends it to A4X Max instances. The post gives no performance figures, prices or precise availability dates.
Summary written by AnnuIA in its own words from the official page of NVIDIA. The content and rights belong to the publisher: please read the original article.
More announcements from NVIDIA
- Agentic AI joins the search for extraterrestrial signals
- Thin, light and built for everything: NVIDIA RTX Spark Windows laptops gain new models, agent tools and games
- How AI is changing the pace of discovery for millions living with dementia
- Leading European financial services organizations embrace open models to build proprietary intelligence