The integration couples NVIDIA AI Cluster Runtime, which establishes locked deployment recipes for GPU platform components, with Google's k8s-aibom tool to monitor active workloads higher up the software stack. Emitting CycloneDX 1.6 reports, the utility catalogs inference runtimes, models, and retrieval-augmented generation dependencies running across Kubernetes nodes. This runtime tracking is enabled by default in two standard Google Kubernetes Engine recipes designed for NVIDIA H100 GPUs running Container-Optimized OS.
Key points
Pairing static configuration validation with active runtime inventory mitigates blind spo…: Pairing static configuration validation with active runtime inventory mitigates blind spots caused by mutable container tags.
The system automatically detects shadow workloads and configuration drift across workload…: The system automatically detects shadow workloads and configuration drift across workloads like NIMService and LeaderWorkerSet.
NVIDIA validates the security of the integrated tool by verifying provenance against Goog…: NVIDIA validates the security of the integrated tool by verifying provenance against Google's build identity alongside role-based controls.
Sources checked by our analysis pipeline: 1
As teams deploy and scale AI workloads using NVIDIA Dynamo, Kubeflow, or Slurm, tracking what actually runs requires visibility beyond the platform configuration captured in a recipe. Application teams define and update these workloads independently of the platform recipe, so a reproducible deployment bundle alone cannot establish which workload artifacts are running on each node. Mutable image tags like :latest
add another layer of complexity. Two nodes can end up running different code under identical tags, and a node that pulls next week may retrieve an image that was not covered by an earlier security review.
NVIDIA AI Cluster Runtime (AICR), an open source modular component of the NVIDIA DSX AI Factory Platform, now includes a qualified AI bill of materials (AIBOM) component to close that gap. An open source project created by Google, k8s-aibom tracks AI workloads running on Kubernetes. Two stock recipes select k8s-aibom by default: h100-gke-cos-inference
and h100-gke-cos-inference-dynamo
. Other GKE recipes can opt in during generation with --runtime-inventory
enabled. Recipes selecting it can opt out with --runtime-inventory
disabled.
AICR turns known-good GPU platform configurations into version-locked…
Cultural organizations are accelerating digital preservation initiatives to make historic archives and Indigenous films accessible to wider international audiences. Concurrently, the technology sector is dealing with critical infrastructure updates and software security vulnerabilities across major platforms. Meanwhile, social media operators face mounting scrutiny over algorithmic harms and content moderation failures that expose users to safety risks.
Developer Bo Lau has launched an interactive project titled "Term Sheet, Please" that reimagines venture capital screening through document-inspection simulation mechanics. Set immediately in the wake of Demo Day, the narrative-driven game casts players as investors scrutinizing applicant materials to make high-stakes financial calls. This simulation gamifies the startup evaluation process by requiring users to conduct due diligence directly from an office desk.
Tech leaders who sought common ground with Donald Trump now find themselves in a precarious position across multiple political and competitive fronts. Retaliatory measures from Democrats loom as a significant threat, even as public skepticism toward major technology platforms intensifies nationwide. These domestic political vulnerabilities are compounding international risks, particularly as foreign rivals challenge American dominance in emerging software sectors.