New Gaslight macOS Malware Uses Prompt Injection to Disrupt AI-Assisted Analysis
Summary: SentinelOne researchers documented a previously undocumented Rust-based macOS implant and information stealer, codenamed Gaslight, that they assess with high confidence is the work of North Korea-aligned threat actors. The malware uses a Telegram bot API command-and-control channel with a polling loop and six shell commands, persists via a LaunchAgent labeled "com.apple.system.services.activity," and embeds a 6.6 KB Base64-encoded Python script that harvests Terminal histories, installed applications, running processes, system profiles, the macOS Keychain database, and data from Chrome, Brave, Firefox, and Safari before zipping it to "temp/collected_data.zip" and uploading it via Telegram.
Bottom line: Gaslight carries 38 fabricated "system" messages - fake token expiry, out-of-memory kills, disk exhaustion, and bogus injection warnings - built to make an LLM-assisted triage agent abort or refuse analysis, an attempt to weaponize the reverse-engineering pipelines that increasingly sit in front of analyst judgment. The implant also self-redacts its Telegram bot token at runtime, denying it to anyone who captures logs or crash artifacts.
Read the original at The Hacker News ↗
- Topics
- AI Security
- Related
- SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 8 Cyber Risks · Nov 15, 2026Pentagon Renames CDAO ‘Super Intelligence Office’ Cyber Risks · Nov 15, 2026186 Fake AI Domains Redirect Users to Click Fix Lures, Browser Malware, and Search Hijackers Cyber Risks · Nov 12, 2026