224 matches found
ONNX download_model_with_test_data() Code Execution
The vulnerability of the downloadmodelwithtestdata function in the software library for building deep learning neural networks, Open Neural Network Exchange ONNX, is related to an incorrect path name limitation for accessing restricted directories. Exploiting this vulnerability could allow a...
Open Neural Network Exchange Path Traversal Vulnerability
Open Neural Network Exchange ONNX is an open ecosystem that enables AI developers to choose the right tools as their projects evolve. A path traversal vulnerability exists in Open Neural Network Exchange version 1.16.0, which stems from insufficient protection against path traversal attacks in...
Open Neural Network Exchange Path Traversal Vulnerability
Open Neural Network Exchange ONNX is an open ecosystem that enables AI developers to choose the right tools as their projects evolve. A security vulnerability exists in Open Neural Network Exchange versions 1.15.0 and earlier, which stems from a field in the externaldata tensor prototype that may...
Open Neural Network Exchange Buffer Error Vulnerability
The Open Neural Network Exchange ONNX is an open ecosystem that enables AI developers to choose the right tools for their projects as they evolve. A buffer error vulnerability exists in Open Neural Network Exchange 1.15.0 and prior versions, which stems from the ONNXASSERT and ONNXASSERTM functio...
Security Bulletin: Mutiple Vulnerabilties in Open Source packages affecting IBM Watson Machine Learning Accelerator on Cloud Pak for Data
Summary IBM Watson Machine Learning Accelerator on Cloud Pak for Data is vulnerable to several open source vulnerabilites. This bulletin identifies the steps to take to address the vulnerabilities. Vulnerability Details CVEID:CVE-2022-1996 DESCRIPTION: go-restful could allow a remote attacker to...
SUSE CVE-2022-26076
Uncontrolled search path element in the IntelR oneAPI Deep Neural Network oneDNN before version 2022.1 may allow an authenticated user to potentially enable escalation of privilege via local access...
CVE-2022-26076
Uncontrolled search path element in the IntelR oneAPI Deep Neural Network oneDNN before version 2022.1 may allow an authenticated user to potentially enable escalation of privilege via local access...
UBUNTU-CVE-2022-26076
Uncontrolled search path element in the IntelR oneAPI Deep Neural Network oneDNN before version 2022.1 may allow an authenticated user to potentially enable escalation of privilege via local access...
CVE-2022-26076
Uncontrolled search path element in the IntelR oneAPI Deep Neural Network oneDNN before version 2022.1 may allow an authenticated user to potentially enable escalation of privilege via local access...
CVE-2022-26076
Intel oneAPI Deep Neural Network (oneDNN) before 2022.1 contains an uncontrolled search path element that may allow an authenticated user to escalate privileges via local access. Public sources (Red Hat CVE entry and Intel advisory) confirm the affected component and the local-privilege escalatio...
Intel OneApi Toolkits 代码问题漏洞
Intel OneApi Toolkits is a set of core tools and libraries from the United States Intel Intel. for developing high-performance, data-centric applications across different architectures. A security vulnerability exists in Intel oneAPI Deep Neural Network oneDNN versions prior to 2022.1, which stem...
CVE-2022-26076: Uncontrolled Search Path Element
Uncontrolled search path element in the IntelR oneAPI Deep Neural Network oneDNN before version 2022.1 may allow an authenticated user to potentially enable escalation of privilege via local access...
IoC detection experiments with ChatGPT
ChatGPT is a groundbreaking chatbot powered by the neural network-based language model text-davinci-003 and trained on a large dataset of text from the Internet. It is capable of generating human-like text in a wide range of styles and formats. ChatGPT can be fine-tuned for specific tasks, such a...
PYSEC-2023-38
Versions of the package onnx before 1.13.0 are vulnerable to Directory Traversal as the externaldata field of the tensor proto can have a path to the file which is outside the model current directory or user-provided directory, for example "../../../etc/passwd"...
Open Neural Network Exchange 路径遍历漏洞
The Open Neural Network Exchange ONNX is an open ecosystem that enables AI developers to choose the right tools as their projects evolve. A security vulnerability exists in Open Neural Network Exchange version 1.13.0, which stems from the fact that the externaldata field of a tensor proto may poi...
Sony Neural Network Libraries reliance on untrusted inputs prior to v1.0.10
nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla prior to v1.0.10 relies on the HOME environment variable, which might be untrusted...
GHSA-4Q2W-RW7M-XQW6 Sony Neural Network Libraries reliance on untrusted inputs prior to v1.0.10
nbla/logger.cpp in libnnabla.a in Sony Neural Network Libraries aka nnabla prior to v1.0.10 relies on the HOME environment variable, which might be untrusted...
Using “Master Faces” to Bypass Face-Recognition Authenticating Systems
Fascinating research: "Generating Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution." Abstract: A master face is a face image that passes face-based identity-authentication for a large portion of the population. These faces can be used to impersonate, with a high...
Hiding Malware in ML Models
Interesting research: "EvilModel: Hiding Malware Inside of Neural Network Models". Abstract: Delivering malware covertly and detection-evadingly is critical to advanced malware campaigns. In this paper, we present a method that delivers malware covertly and detection-evadingly through neural...
In0ri - Defacement Detection With Deep Learning
In0ri is a defacement detection system utilizing a image-classification convolutional neural network. Introduction When monitoring a website, In0ri will periodically take a screenshot of the website then put it through a preprocessor that will resize the image down to 250x250px and numericalize t...