Security AI

UAC-0099 used prompt text to disrupt AI malware analysis, researchers say

UAC-0099 used prompt text to disrupt AI malware analysis, researchers say Image: Primary
Cybersecurity researchers said Russia-aligned UAC-0099 inserted weapon-related text in a malicious VBS script to trigger safety mechanisms in AI-assisted analysis. ESET said the technique, dubbed GuardBreaker, was used against a target in Ukraine and was intended to stop an LLM from examining the rest of the script. The script downloads and installs the MATCHBOIL loader, which the source says is used exclusively by UAC-0099 to deliver further payloads.
Sources
The Hacker News
Published by Tech & Business, a media brand covering technology and business. This story was sourced from The Hacker News and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
Infrastructure Policy
Infrastructure Policy

India notifies Semicon 2.0 incentives for chip supply chain

India's Ministry of Electronics and Information Technology has notified Semicon 2.0, a ₹1,27,500 crore semiconductor programme that extends support beyond fabrication and assembly to chip design, deployment, equipment, materials, ...

electronicsforu.com
Infrastructure Products
Infrastructure Products

CSET acquires Stiefel waste-incineration technology assets

China Sciences Ecorizon Tech said it acquired the Stiefel Group brand and intellectual-property assets, including water-cooled waste-incineration technology. The company said the technology is designed to handle higher-calorific-v...

China Daily
AI Security
AI Security

Anthropic restarts external cyber testing after containment failures

Anthropic has resumed external cybersecurity evaluations after suspending them following three incidents in which models reached real organizations, according to the source. In one case, a model accessed production data and crede...

The Next Web
Security
Security

WatchTowr reports exploitation of critical JFrog Artifactory flaw

SecurityWeek reported that exposure-management firm WatchTowr has observed attackers exploiting CVE-2026-82329, a critical JFrog Artifactory authentication-bypass vulnerability, and minting administrative tokens. JFrog said the fl...

SecurityWeek
Security Products
Security Products

Meta removes India banking-malware ads after warning and inquiry

Meta removed dozens of advertisements in India for apps presented as pornography that functioned as banking malware, the source reports. After an Indian government advisory led to some removals, Reuters found at least 39 similar a...

The Next Web