Chinese AI company Moonshot AI reportedly discussed renting Nvidia Blackwell computing resources from a European neocloud (a cloud provider dedicated to AI computing). Because the chips are used remotely rather than physically moved, the arrangement has become a regulatory issue.
NVIDIA has built a system in which multiple AI agents work together to help run its own AI data centers. The agents find problems, connect the causes, and show recommendations to the responsible engineer through chat. The goal is to reduce the effort of checking multiple dashboards one by one.
NVIDIA has officially released CUDA 13.5, its toolkit for developing GPU programs. The next-generation Vera Rubin GPU architecture can now be used in real production environments, and Windows on Arm support now extends to the DGX Station Windows version.
NVIDIA has introduced the FOX Blueprint, a reference design for building AI agents that manage factories. Multiple specialized agents work together to inspect defects, assign repairs, and investigate causes. It can also be installed so that it runs only inside a factory's internal network.
NVIDIA explains how to analyze the failure records of AI agents to find areas for improvement, turn those findings into tests, and fix the agent's setup. The process decides what to fix first based on the failure records. After fixes, the agent is tested on new and existing tasks to confirm it has improved.
Chinese auto parts maker Pateo has put AI language models from NVIDIA and ByteDance into an AI device (AI Box) that runs inside cars. The key goal is to allow voice conversation and control of car functions inside the vehicle, even when the internet is down.
Chinese auto parts company Pateo is upgrading in-car AI by adding Nvidia and ByteDance language models to its vehicle AI box, which is built on Nvidia's high-performance chips. The system is designed to run AI right inside the car, even when the internet is down.
China's auto parts company Pateo is upgrading in-car AI by adding NVIDIA and ByteDance language models to an AI box based on NVIDIA chips. The key goal is to let AI work right inside the car even when the internet is cut off.
China's Pateo is making AI in cars smarter by adding voice recognition and large language models to an AI device built on NVIDIA's in-vehicle computer. Some functions can run directly in the car even when the internet is weak.
Chinese auto parts maker Pateo is strengthening in-car AI features by adding two language models to a vehicle AI box built on Nvidia chips. The key feature is that the AI can run right inside the car, even when there is no internet connection.
Chinese auto parts maker Pateo is upgrading in-car AI by adding two language models to an AI box built on Nvidia's high-performance automotive chip. A key feature is that the AI runs right inside the car even when there is no internet connection.
This is a telecom session video from NVIDIA's GTC Berlin event. It covers how telecom companies will link AI data centers, edge AI (running AI close to devices), and AI-RAN (using AI to run wireless networks) on the way to 6G.
Cities collect large amounts of data from cameras and sensors, but turning that data into action is difficult. Nvidia and its partners are expanding examples in several cities that use physical AI and AI agents to close this gap. The key is letting local governments keep direct control over sensitive data while quickly improving public services.