Kraken: Higher-Order EM Side-Channel Attacks on DNNs in near and Far Field

The multi-million dollar investment required for modern machine learning ML has made large ML models a prime target for theft. In response, the field of model stealing has emerged. Attacks based on physical side-channel information have shown that DNN model extraction is feasible, even on CUDA...

SeBERTis: A Framework for Producing Classifiers of Security-Related Issue Reports

Monitoring issue tracker submissions is a crucial software maintenance activity. A key goal is the prioritization of high risk, security-related bugs. If such bugs can be recognized early, the risk of propagation to dependent products and endangerment of stakeholder benefits can be mitigated. To...

Binary and Multiclass Cyberattack Classification on GeNIS Dataset

The integration of Artificial Intelligence AI in Network Intrusion Detection Systems NIDS is a promising approach to tackle the increasing sophistication of cyberattacks. However, since Machine Learning ML and Deep Learning DL models rely heavily on the quality of their training data, the lack of...

Exploring the Effect of DNN Depth on Adversarial Attacks in Network Intrusion Detection Systems

Adversarial attacks pose significant challenges to Machine Learning ML systems and especially Deep Neural Networks DNNs by subtly manipulating inputs to induce incorrect predictions. This paper investigates whether increasing the layer depth of deep neural networks affects their robustness agains...

The vulnerability of the FortiMail email security system, a software-hardware solution for information protection based on AI and deep neural networks from Fortinet’s FortiNDR (Network Detection and Response), arises from the possibility of copying buffers without checking the size of the input data. This allows attackers to execute arbitrary code.

The vulnerability of the FortiMail email security system, a software-hardware solution for information protection based on AI and deep neural networks from Fortinet, is related to the copying of buffers without checking the size of the input data. Exploiting this vulnerability allows an attacker...

Technical Evaluation of a Disruptive Approach in Homomorphic AI

We present a technical evaluation of a new, disruptive cryptographic approach to data security, known as HbHAI Hash-based Homomorphic Artificial Intelligence. HbHAI is based on a novel class of key-dependent hash functions that naturally preserve most similarity properties, most AI algorithms rel...

Learning from the Good Ones: Risk Profiling-Based Defenses against Evasion Attacks on DNNs

Safety-critical applications such as healthcare and autonomous vehicles use deep neural networks DNN to make predictions and infer decisions. DNNs are susceptible to evasion attacks, where an adversary crafts a malicious data instance to trick the DNN into making wrong decisions at inference time...

Cert-SSB: toward Certified Sample-Specific Backdoor Defense

Deep neural networks DNNs are vulnerable to backdoor attacks, where an attacker manipulates a small portion of the training data to implant hidden backdoors into the model. The compromised model behaves normally on clean samples but misclassifies backdoored samples into the attacker-specified...

BIT-PYTORCH-2025-32434 PyTorch: `torch.load` with `weights_only=True` leads to remote code execution

PyTorch is a Python package that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system. In version 2.5.1 and prior, a Remote Command Execution RCE vulnerability exists in PyTorch when loading a model using torch.load with...

Implementing Cryptography in AI Systems

Interesting research: "How to Securely Implement Cryptography in Deep Neural Networks." Abstract: The wide adoption of deep neural networks DNNs raises the question of how can we equip them with a desired cryptographic functionality e.g, to decrypt an encrypted input, to verify that this input is...

Manipulating Weights in Face-Recognition AI Systems

Interesting research: "Facial Misrecognition Systems: Simple Weight Manipulations Force DNNs to Err Only on Specific Persons": Abstract: In this paper we describe how to plant novel types of backdoors in any facial recognition model based on the popular architecture of deep Siamese neural network...

Black Hat: Scaling Automated Disinformation for Misery and Profit

LAS VEGAS – Researchers recently demonstrated the weaponization of deep neural networks that can be used to shape public opinion, enrage people on Twitter and possibly spark QAnon 2.0. The research, presented last week at Black Hat by Drew Lohn, senior fellow at the Center for Security and...

Locating malicious drone operators through deep neural networks

By Zara Khan Researchers at Ben Gurion University have developed a technique... This is a post from HackRead.com Read the original post: Locating malicious drone operators through deep neural networks...

deep-pwning - Metasploit for Machine Learning

Deep-pwning is a lightweight framework for experimenting with machine learning models with the goal of evaluating their robustness against a motivated adversary. Note that deep-pwning in its current state is no where close to maturity or completion. It is meant to be experimented with, expanded...

Free Open Source Face Recognition Neural Network: OpenFace

OpenFace is a Python and Torch implementation of face recognition with deep neural networks and is based on the CVPR 2015 paper FaceNet: A Unified Embedding for Face Recognition and Clustering by Florian Schroff, Dmitry Kalenichenko, and James Philbin at Google. Torch allows the network to be...