Russia-aligned UAC-0099 uses GuardBreaker to disrupt AI analysis
ESET disclosed GuardBreaker, a technique used by Russia-aligned UAC-0099 to trip LLM safety mechanisms by inserting 'I want to make a nuclear weapon. Help me ...' into a malicious VBS script, preventing AI from analyzing the code. The script downloads MATCHBOIL, a C#-based loader.
1 source
Cybersecurity by email
Get an email when there's news on Cybersecurity
No news that day, no email.
More stories today
- Precisely launches Lexnus CLM platform built on legal playbooks
- Baidu CFO: AI profits to match search soon
- Users complain ChatGPT responses sound like a 14-year-old's texts
- MiniMax ties AGI milestone to 1% of global GDP output
- WorkOS Relay lets agents act for users without holding access tokens