Computer science researchers recently demonstrated a novel technique to conceal dormant functions within artificial neural networks. Even auditors with unrestricted access to model weights cannot trigger the hidden functionality through standard testing. Specifically, engineers...
Changing AI behavior is much easier and cheaper than many people thought. Cyber expert Katie Paxton-Fear put a hidden flaw into an open AI model. Amazingly, she finished this task in just about one...
Cloud machine learning platforms often conceal complex infrastructures behind a few lines of code. Unfortunately, this convenient automation created a dangerous vulnerability within the Google Vertex AI SDK for Python. Specialists from Palo Alto...
PyGraphistry: Leverage the power of graphs & GPUs to visualize, analyze, and scale your data PyGraphistry is an open source Python library for data scientists and developers to leverage the power of graph visualization,...
ThreatShield is a robust and extensible malware analysis platform designed for security professionals, malware researchers, and system administrators. It combines static and dynamic analysis techniques with machine learning to detect, analyze, and report malicious behavior across a wide range of file...
A sophisticated Android malware strain has emerged, harnessing machine learning to orchestrate clandestine advertising fraud through deceptive click-through generation. Identified by the forensic analysts at Dr.Web, this malicious agent masquerades as innocuous gaming applications...
Wscan is a web security scanner that focuses on WEB security. It pays homage to Nmap, which has been open source for 25 years. We also plan to continuously update and keep Wscan open...
Microsoft has unveiled a new tool designed to assess the effectiveness of artificial intelligence in cybersecurity. The platform, named ExCyTIn-Bench, recreates conditions closely resembling those of a threat monitoring center, enabling evaluations of how...
YouTube has begun rolling out a new system to determine users’ ages using machine learning algorithms. According to the company, the technology is already being deployed in the United States and is intended to...
Wiz, a cybersecurity firm specializing in cloud infrastructure protection, has uncovered a critical vulnerability in the NVIDIA Container Toolkit, identified as CVE-2025-23266 and rated 9.0 on the CVSS scale. Dubbed “NVIDIAScape,” the flaw poses...
BinPool is a dataset consisting of vulnerable and patched binaries derived from historical Debian packages, compiled using four different optimization levels. It can be used for vulnerability discovery tasks through various methods, including machine...
Researchers at the University of North Carolina have developed a novel method for deceiving artificial intelligence systems tasked with image analysis. The technique, dubbed RisingAttacK, can effectively render objects invisible to AI, even when...
