Preface: The security concerns regarding data integrity in Peripheral Component Interconnect Express (PCIe) transactions are a critical aspect of modern computing, particularly in data centers, cloud environments, and automotive systems where sensitive information is routinely handled. Historically, PCIe interfaces were considered relatively secure due to their placement inside a server enclosure, but the rise of disaggregated systems, confidential computing, and sophisticated physical attacks has changed this perspective entirely. As an interconnect that links the CPU to various peripherals like GPUs, SSDs, and network adapters, any vulnerability can have far-reaching consequences, leading to data corruption, unauthorized access, or system compromise.

Background: AMD EPYC processors use an I/O Die (IOD) to manage all external interfaces, connecting to CPU Dies (CCDs) via high-speed Global Memory Interconnect (GMI) links and handling numerous DDR5 memory channels, PCIe Gen5, and CXL lanes, with SERDES (Serializer/Deserializer) technology underpinning these fast connections for massive bandwidth and low latency in data-intensive workloads, allowing for up to 12 memory channels and 128 PCIe lanes per socket in recent generations.

AMD SERDES technology significantly enhances the physical-layer data integrity and signal quality in PCIe transactions, but it is distinct from higher-level security features like encryption. SERDES technology is a foundational element that ensures reliable data transmission at extremely high speeds.

Affected Products and Mitigation:

From security point of view, it expect additional details from the PCIe SIG and plan to update this security notice as more information is available.  At this time, AMD believes the following products may be impacted.

AMD EPYC™ 9005 Series Processors

AMD EPYC™ Embedded 9005 Series Processors

Ref: PCI-SIG (Peripheral Component Interconnect Special Interest Group) is the electronics industry consortium that defines and maintains the standards for PCI, PCI-X, and PCI Express (PCIe) computer buses, ensuring high-speed, interoperable connectivity for components like graphics cards, SSDs, and network adapters in computers and data centers. This non-profit group, with hundreds of member companies, develops specifications, promotes compliance, and fosters an open ecosystem for PCIe technology, allowing different manufacturers’ products to work together seamlessly.

Vulnerability details:

CVE-2025-9612 (non-AMD) : An issue was discovered in the PCI Express (PCIe) Integrity and Data Encryption (IDE) specification, where insufficient guidance on Transaction Layer Packet (TLP) ordering and tag uniqueness may allow encrypted packets to be replayed or reordered without detection. This can enable local or physical attackers on the PCIe bus to violate data integrity protections.

CVE-2025-9613 (non-AMD): A vulnerability was discovered in the PCI Express (PCIe) Integrity and Data Encryption (IDE) specification, where insufficient guidance on tag reuse after completion timeouts may allow multiple outstanding Non-Posted Requests to share the same tag. This tag aliasing condition can result in completions being delivered to the wrong security context, potentially compromising data integrity and confidentiality.

CVE-2025-9614 (non-AMD): An issue was discovered in the PCI Express (PCIe) Integrity and Data Encryption (IDE) specification, where insufficient guidance on re-keying and stream flushing during device rebinding may allow stale write transactions from a previous security context to be processed in a new one. This can lead to unintended data access across trusted domains, compromising confidentiality and integrity.

Official announcement: Please refer to the link for details –

https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7056.html

Initial release – 02 December 2025   

Preface: Triton Inference Server is an open source inference serving software that streamlines AI inferencing. Triton Inference Server enables teams to deploy any AI model from multiple deep learning and machine learning frameworks, including TensorRT, PyTorch, ONNX, OpenVINO, Python, RAPIDS FIL, and more.

Background: The NVIDIA Triton Inference Server natively supports gRPC as one of its primary communication protocols for the client API. Furthermore, gRPC can also be used for health checks, statistics, and model loading/unloading operations, not just inference requests. Inference requests arrive at the server via either HTTP/REST or GRPC or by the C API and are then routed to the appropriate per-model scheduler. Triton implements multiple scheduling and batching algorithms that can be configured on a model-by-model basis.

Triton major features:

• Supports multiple deep learning frameworks
• Supports multiple machine learning frameworks
• Concurrent model execution
• Dynamic batching
• Sequence batching and implicit state management for stateful models
• Provides Backend API that allows adding custom backends and pre/post processing operations
• Model pipelines using Ensembling or Business Logic Scripting (BLS)
• HTTP/REST and GRPC inference protocols based on the community developed KServe protocol
• A C API and Java API allow Triton to link directly into your application for edge and other in-process use cases
• Metrics indicating GPU utilization, server throughput, server latency, and more

Vulnerability details:

CVE-2025-33211         NVIDIA Triton Server for Linux contains a vulnerability where an attacker may cause an improper validation of specified quantity in input. A successful exploit of this vulnerability may lead to denial of service.

CVE-2025-33201         NVIDIA Triton Inference Server contains a vulnerability where an attacker may cause an improper check for unusual or exceptional conditions issue by sending extra large payloads. A successful exploit of this vulnerability may lead to denial of service.

Official announcement: Please refer to the link for details –

https://nvidia.custhelp.com/app/answers/detail/a_id/5734

Official Updated 11/26/2025

Preface: AI vision models are artificial intelligence systems, often multimodal (Vision-Language Models or VLMs), that analyze and interpret visual data (images, videos) alongside text, enabling them to “see” and understand the world like humans, performing tasks from object recognition and image captioning to answering questions about visuals and generating new images, by converting visual info into a format comparable to text.

Background: You use NVIDIA TAO (Train, Adapt, Optimize) to rapidly build, customize, and deploy high-performance, domain-specific AI models (especially for vision) with less code, less data, and faster training by leveraging powerful pre-trained foundation models, fine-tuning them with your own data, and optimizing them for efficient inference on edge-to-cloud devices, saving significant time and resources.

The NVIDIA TAO Toolkit is designed to function with both real and synthetic data.

Training with Real Data: The primary function of the TAO Toolkit is to fine-tune NVIDIA’s extensive library of pretrained foundation models using your own proprietary (real-world) datasets. This process is low-code and enables the customization of models for specific use cases without needing deep AI expertise or training from scratch.

Leveraging Synthetic Data: Synthetic data is often used to address the challenges associated with real data collection, such as scarcity, expensive labeling, and rare edge cases.

Models can be initially trained on large volumes of synthetic data generated from tools like NVIDIA Omniverse Replicator or partner platforms (e.g., Sky Engine AI, AI. Reverie, Lexset).

Vulnerability details: (CVE-2025-33208) NVIDIA TAO contains a vulnerability where an attacker may cause a resource to be loaded via an uncontrolled search path. A successful exploit of this vulnerability may lead to escalation of privileges, data tampering, denial of service, information disclosure

Official announcement: Please refer to the link for more details.

https://nvidia.custhelp.com/app/answers/detail/a_id/5730

Official Update 11/21/2025 04:36 PM

Preface: NeMo Curator is a Python library that includes a suite of modules for data-mining and synthetic data generation. They are scalable and optimized for GPUs, making them ideal for curating natural language data to train or fine-tune LLMs. With NeMo Curator, researchers in Natural Language Processing (NLP) can efficiently extract high-quality text from extensive raw web data sources.

NVIDIA NeMo Curator, particularly its image curation modules, requires a CUDA-enabled NVIDIA GPU and the corresponding CUDA Toolkit. The CUDA Toolkit is not installed as part of the NeMo Curator installation process itself, but rather is a prerequisite for utilizing GPU-accelerated features.

Background: NeMo Framework includes NeMo Curator because high-quality data is essential for training accurate generative AI models, and Curator provides a scalable, GPU-accelerated toolset for processing and preparing large datasets efficiently. It handles everything from cleaning and deduplicating text to generating synthetic data for model customization and evaluation, preventing data processing from becoming a bottleneck.

Potential risks under observation: The vulnerability arises when malicious files—such as JSONL files—are loaded by NeMo Curator. If these files are crafted to exploit weaknesses in how NeMo Curator parses or processes them, they can inject executable code.

Ref: Parser is related to predefined variables, as it can either parse data into variables or use predefined variables to perform its task.

Vulnerability details:

CVE-2025-33204         NVIDIA NeMo Framework for all platforms contains a vulnerability in the NLP and LLM components, where malicious data created by an attacker could cause code injection. A successful exploit of this vulnerability may lead to code execution, escalation of privileges, information disclosure, and data tampering.

CVE-2025-33205         NVIDIA NeMo framework contains a vulnerability in a predefined variable, where an attacker could cause inclusion of functionality from an untrusted control sphere by use of a predefined variable. A successful exploit of this vulnerability may lead to code execution.

Official announcement: Please refer to the link for details –

https://nvidia.custhelp.com/app/answers/detail/a_id/5729