All news with #nvidia tag

🔒 A vulnerability in Nvidia's DCGM Exporter telemetry agent could allow unauthenticated resource exhaustion and information disclosure. The flaw, tracked as CVE-2026-47483, lets attackers crash the monitoring service by abusing the /debug/pprof/ endpoints, potentially disrupting AI workloads on the same host. Nvidia released a patched DCGM Exporter version 4.8.2 and Lava Security recommended restricting public access and disabling profiling where not needed.

🔎 Since launching SynthID in 2023, Google has embedded imperceptible watermarks into billions of images and videos and hundreds of thousands of years of audio to help identify AI-generated media. The company previously offered an early SynthID Detector for media professionals; today it expands access globally in English. The detector can identify content produced by Google and partner models such as OpenAI, NVIDIA, and Kakao, with more partners planned. This complements existing verification features across Search, Gemini, and Chrome.

🖥️ Amazon WorkSpaces Core Managed Instances now support Graphics G7 instances powered by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs and Intel Xeon 6 processors. G7 offers up to 2.1X better performance than G6 for graphics-intensive workloads and includes 32 GB of GDDR7 per GPU with 2.67X faster memory bandwidth. Six sizes are available (1–8 GPUs, 8–192 vCPUs, 32–768 GB RAM) with Linux and Windows support and BYOL.

📄 The White House has obtained a voluntary safety commitment from six leading AI firms, who agreed to internal controls, independent audits and board-level oversight for frontier models. President Trump and the executives signed the White House Accord on Super Intelligence on September 29. Signatories include leaders from Google, Anthropic, Meta, OpenAI, xAI and NVIDIA. The accord outlines four layers of controls and calls for regular meetings to develop standards and best practices.

🔒 Nvidia unveiled the Open Agent Safety Platform, combining OpenShell software with DPU-based silicon in a reference design to secure agentic AI from testing through deployment. The platform uses NVIDIA Sentry on BlueField-4 DPUs to monitor and enforce policies out-of-band, while OpenShell provides a secure runtime boundary on Vera CPUs and can be extended to other platforms. Partners include major vendors and financial firms, though notable hyperscalers are absent. Analysts praise the hardware-enforced controls but warn they only cover agents running inside the governed runtime and cannot address unknown or external agents.

🔒 NVIDIA has introduced an Open Agent Safety Platform to enforce controls on autonomous AI agents throughout testing and deployment. The platform combines OpenShell, an open-source runtime that traces agent activity and enforces policies, with Sentry, a hardware watchdog that can quarantine agents in milliseconds. Announced on September 28, the platform aims to add enforcement at the hardware and compute layers in addition to model-level controls. More than 100 organizations and major vendors are already collaborating on the effort.

🔒 This article examines how NVIDIA Open Agent Safety Platform and Check Point's semantic monitoring address gaps in autonomous agent control. It explains that prompts and model safeguards can be circumvented, so OpenShell enforces an infrastructure-level boundary while NVIDIA Sentry provides out-of-band hardware-isolated enforcement. Check Point integrates semantic monitoring to judge whether successive actions still match the agent's task and enforces controls through OpenShell's middleware.

🔒 Palo Alto Networks and NVIDIA detail a joint architecture to secure autonomous AI agents by combining NVIDIA’s accelerated compute and secure runtime with Palo Alto Networks’ Prisma AIRS and IDIRA capabilities. The integration places governance and enforcement at the infrastructure layer, using Prisma AIRS AI Gateway on NVIDIA Vera CPUs and Prisma AIRS AI Runtime Security on NVIDIA BlueField DPUs. The design emphasizes continuous identity verification, least-privilege access, data redaction, and deep inspection to prevent unauthorized actions and privilege escalation.

🚀 Amazon WorkSpaces Personal and Core now offer Graphics G7 bundles powered by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs and Intel Xeon 6 processors. These bundles deliver up to 2.1x better graphics performance versus Graphics G6 and come in four sizes from 8 vCPUs/32 GB/1 GPU to 48 vCPUs/192 GB/2 GPUs. Graphics G7 supports demanding professional workloads, Windows licensing options, AlwaysOn and AutoStop modes, and is initially available in three US Regions with broader rollout planned.

🚀 Amazon SageMaker JumpStart now includes NVIDIA's Qwen3.6-35B-A3B-NVFP4 and Alibaba's Wan2.1-T2V-1.3B-Diffusers models, expanding available foundation models for AWS customers. Qwen3.6-35B-A3B-NVFP4 is a Mixture-of-Experts model optimized for agentic coding, multimodal and long-context reasoning, quantized to NVFP4 to reduce memory footprint while supporting very long context windows. Wan2.1-T2V-1.3B-Diffusers targets lightweight text-to-video generation, delivering 480p clips on consumer GPUs with modest VRAM requirements. Deployments are available via the SageMaker JumpStart model catalog or the SageMaker Python SDK.

📣 Amazon SageMaker JumpStart now offers Google DeepMind’s Gemma-4-31B-it-assistant and NVIDIA-quantized Gemma-4-31B-IT-NVFP4, bringing the Gemma 4 31B dense architecture to enterprise workloads in full-precision and optimized 4-bit FP4 variants. The assistant-tuned model supports multimodal reasoning, large 256K-token contexts, and native function calling, while the NVFP4 variant reduces memory footprint and speeds inference for cost-efficient production. Deployments are available via the SageMaker console or Python SDK.

🚀 Google Cloud announces enhancements to Dataflow to support large AI workloads, including GA of Pause/Resume for batch jobs and support for G4 VMs with NVIDIA RTX PRO 6000 Blackwell GPUs. Pause/Resume reduces wasted compute by enabling stopped batch jobs to be resumed, improving productivity and resource utilization. The new GPU support delivers larger memory and bandwidth for in-pipeline inference using models of 70B+ parameters while preserving native RunInference and autoscaling capabilities.

🔍 A University of Toronto paper describes GPUThor, an advanced Rowhammer-style attack targeting GDDR6 video memory on Nvidia Ampere accelerators. The researchers show a novel access pattern that defeats TRR mitigation by exploiting its refresh cadence, producing far more bit flips than prior GPU attacks. Results include large numbers of multi-bit errors and observed denial-of-service effects, though arbitrary code execution remains unproven. The work highlights ongoing risks to shared GPU infrastructure used in cloud and AI workloads.

🚀 Starting today, Amazon EC2 P6-B200 instances accelerated by NVIDIA Blackwell GPUs are available in the AWS Asia Pacific (Hyderabad) Region. These instances deliver up to 2x performance versus P5en for AI training and inference and include 8 Blackwell GPUs, 1440 GB of high-bandwidth GPU memory, and 60% greater GPU memory bandwidth. They use 5th Gen Intel Xeon processors (Emerald Rapids), offer up to 3.2 Tbps EFAv4 networking, and run on the AWS Nitro System for scalable UltraClusters.

🚀 Starting today, Amazon EC2 P6-B300 instances are available in the AWS Asia Pacific (Jakarta) Region. These instances provide 8x NVIDIA Blackwell Ultra GPUs with 2.1 TB high-bandwidth GPU memory, 6.4 Tbps EFA networking, 300 Gbps ENA throughput, and 4 TB system memory. P6-B300 delivers higher networking and memory than P6-B200, enabling faster training and greater token throughput for large foundation models.

🖥️ Amazon WorkSpaces Applications now supports Graphics G7 instances powered by NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs and Intel Xeon Scalable (6th Gen) processors. G7 delivers up to 2.1× better performance for graphics workloads versus G6 and offers 32 GB of GDDR7 per GPU with 2.67× faster memory bandwidth. Six sizes are available (1–8 GPUs, 8–192 vCPUs, 32–768 GB RAM) and G7 is initially available in three US regions.

🚀 Amazon EC2 P6-B300 instances are now available in Asia Pacific (Hyderabad) and South America (Sao Paulo), expanding regional availability. These instances offer 8x NVIDIA Blackwell Ultra GPUs with 2.1 TB GPU memory, 6.4 Tbps EFA networking, 300 Gbps ENA throughput, and 4 TB system memory. P6-B300 delivers increased networking, GPU memory, and TFLOPS versus P6-B200, targeting training and deployment of large trillion-parameter FMs and LLMs. The p6-b300.48xlarge size is available in multiple AWS Regions including US West (Oregon) and N. Virginia.

🚀 NVIDIA's Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super are now available in Amazon SageMaker JumpStart, expanding AWS's foundation model offerings for physical AI. These omnimodal world models enable robots, autonomous vehicles, and vision AI to perceive, reason, plan, and act. Customers can deploy the models from the SageMaker JumpStart catalog or use the SageMaker Python SDK for integration.

🛡️ Academic researchers disclosed GPUThor, a Rowhammer attack that induces widespread bit flips on NVIDIA Ampere-class workstation GPUs with GDDR6, defeating recommended ECC mitigations and enabling denial-of-service and host privilege escalation. The University of Toronto team hammered DRAM banks for extended periods on multiple RTX A-series cards, producing up to 377,552 flips per gigabyte on an A5000. The exploit requires running an unprivileged CUDA kernel and the researchers advise avoiding cross-tenant GPU sharing, monitoring ECC counters, and restricting untrusted CUDA workloads.

🛡️ Researchers from the University of Toronto disclosed GPUThor, a Rowhammer variant that defeats SECDED ECC on Ampere-class NVIDIA GPUs, enabling DoS and root privilege escalation. The attack achieves far higher bit-flip rates than prior GPU Rowhammer concepts by exploiting undocumented memory request coalescing and TRR behavior. Tested on RTX A4000–A6000 cards, GPUThor produced thousands of flips per GB and demonstrated both device resets and corrupted page tables leading to host root access. NVIDIA issued guidance recommending SYS-ECC, IOMMU/DMA isolation, telemetry monitoring, and restrictions on untrusted workloads.