NVIDIA Introduces Open Agent Safety Platform for AI Agent Isolation
NVIDIA introduced the Open Agent Safety Platform, combining the open-source OpenShell runtime with optional Sentry hardware oversight for BlueField-4.
Tag NVIDIA
NVIDIA Releases Isaac ROS 5.0 with ROS 2 Lyrical and Ubuntu 24.04 Support
NVIDIA’s new major release of its robotics software stack adds support for ROS 2 Lyrical and Ubuntu 24.04, agent skills, and an interface change requiring migration of some existing code.
NVIDIA Launches DSX Ready for AI Factory Power and Cooling
NVIDIA introduced the DSX Ready program to qualify battery energy storage systems and cooling distribution units for AI factories designed around the DSX platform.
Jensen Huang Rejects Predictions of Humanity’s Extinction Due to AI by 2030
Nvidia CEO Jensen Huang told CBS News that he considers the probability of an AI-caused end of the world by 2030 to be zero. He also rejected slowing development and favored applying existing laws.
MLPerf Inference v6.1 Delivers First Preview Results for Vera Rubin NVL72
MLCommons published the MLPerf Inference v6.1 benchmark results. NVIDIA submitted the first preview results for the Vera Rubin NVL72 system for the DeepSeek-R1 and Qwen3-VL 235B models.
Google, NVIDIA and Emerald AI Launch Flexible Data Center Alliance
The new AI Energy Management Alliance aims to promote data centers that adjust their power consumption or use on-site energy resources in response to grid conditions.
DSX MaxLPS Energy Efficiency: NVIDIA Reports Results
At the AI Infra Summit, NVIDIA presented Lambda’s result with the DSX MaxLPS power management system and cited additional highs for the Vera Rubin NVL72 platform. These are company claims without independent reproduction.
Salesforce and NVIDIA Introduce Koa, a Reasoning Model for Agentforce
Salesforce developed Koa with NVIDIA for multi-step CRM tasks in Agentforce. The company post-trained it on synthetic scenarios and says it did not use customer data.
Perplexity Portable Computer Available in Windows App for PCs with RTX GPUs
Perplexity has made the locally oriented Portable Computer agent available in its Windows app. It requires a compatible NVIDIA RTX GPU with at least 24 GB of VRAM.
NVIDIA Expands CUDA-Q with Logical Layer for Fault-Tolerant Quantum Systems
NVIDIA introduced the open-source CUDA-Q Logical layer for modeling logical qubits, error correction and hardware parameters in quantum systems.