Nvidia <span class="t42">picks OpenShell to </span>tame AI gents

Nvidia on Wednesday introduced the OpenShell Reference Architecture, a blueprint built on its OpenShell secure runtime for creating, operating, observing and policing AI agents on Nvidia hardware, with CEO Jensen Huang saying agents can "think on Rubin, act on Vera and be governed on BlueField." The blueprint requires Nvidia Vera Rubin infrastructure running models such as Nemotron, Vera CPUs for agent harnesses and BlueField-4 DPUs supporting Nvidia Sentry security software, and is available to partners including Canonical, SUSE, Red Hat, Amazon Web Services, Google Cloud and Microsoft Azure. Nvidia says BBVA and Santander are reviewing the spec against their security and governance requirements and that SAP is integrating it with its Joule Studio runtime, while OpenShell still lists open security-related issues on GitHub and collects telemetry data by default with an opt-out.

Nvidia <span data-lab-text size desktop="42" class="t42" picks OpenShell to </span tame AI gents The Register https://www.theregister.com Give your agents freedom to do useful work by locking them down Nvidia https://www.machinebrief.com/glossary/nvidia has proposed a new way to keep corporate AI agents in line. Nvidia on Wednesday introduced its OpenShell Reference Architecture https://nvidianews.nvidia.com/news/openshell-reference-architecture . Based on the company's OpenShell https://www.nvidia.com/en-us/ai/openshell/ nv-accordion-7d74ebf581-item-4510a04d93 secure runtime, It's a recipe for creating, operating, observing, and policing AI agents using OpenShell on Nvidia hardware. "Nvidia created OpenShell to give agents the freedom to do useful work within clearly defined boundaries — contained, monitored and under control," CEO Jensen Huang said in a canned statement. "With the OpenShell Reference Architecture, agents can think on Rubin, act on Vera and be governed on BlueField." AI agents are touted as a powerful automation technology. The reality is that they can be unruly, as demonstrated by security testing OpenAI and Anthropic https://www.machinebrief.com/glossary/anthropic conducted, during which their models exceeded their remit and attacked third-party organizations. Developers of existing agents, like Claude https://www.machinebrief.com/glossary/claude Code, Codex, OpenClaw, and OpenCode, therefore include controls that limit access to tools like curl and bash.Devs also create sandboxes to isolate an agentic process from the host, so that if agents go awry they're contained. OpenShell sits a level above that with a runtime that extends sandboxing to govern gateway-brokered credentials, identity and policy boundaries, and eventually routing and workflows too. At least mostly – OpenShell still has a few open security-related issues https://github.com/NVIDIA/OpenShell/issues?q=is%3Aissue+state%3Aopen+security listed on GitHub. To implement OpenShell, organizations need Nvidia hardware. The company's blueprint calls for Nvidia Vera Rubin infrastructure to run models like Nvidia Nemotron, with agent harnesses handled via Vera CPUs, and Nvidia BlueField-4 DPUs aka SmartNICs propping up Nvidia Sentry security software to monitor agent behavior. The AI biz has made this blueprint available to partners capable of powering Linux and Kubernetes infrastructure. The usual suspects are aboard: Canonical, SUSE and Red Hat. Cloud hyperscalers Amazon Web Services, Google Cloud, and Microsoft Azure have also been tapped as vendors for this vision. And of course Nvidia's floor plan applies to agents from various model builders, including Anthropic, Cohere, Google, Meta, OpenAI, Perplexity https://www.machinebrief.com/glossary/perplexity , Reflection, and SpaceX. Nvidia says several banks – Europe’s BBVA and Santander – are perusing the OpenShell Reference Architecture spec to see how it aligns with their security and governance requirements. And the AI biz is not too modest to let slip that SAP is integrating the OpenShell Reference Architecture with its Joule Studio runtime for safer agent operations. It's worth noting that OpenShell collects telemetry data https://docs.nvidia.com/openshell/latest/observability/telemetry by default, "to improve the project," Nvidia insists. The biz allows users to opt-out, with a cautionary note: "Third-party services, model providers, inference https://www.machinebrief.com/glossary/inference endpoints, agents, or tools that you configure and use with OpenShell may have their own terms and privacy practices." ® Get AI news in your inbox Daily digest of what matters in AI.