Back Thehackernews

New Gaslight macOS Malware Uses Prompt Injection to Disrupt AI

Thehackernews •[email protected] (The Hacker News) • June 25, 2026

Learn how to uncover hidden AI use, see what data it can access, map every AI action to a human owner, and apply practical governance without heavy infrastructure changes.

Learn how to contain Mythos-style AI attacks with practical Zero Trust controls that reduce exposure, stop lateral movement, and limit risk.

The full story

This article is one source in a clustered incident — the cluster page carries the summary, timeline and every other outlet covering it.

New macOS Gaslight Malware Targets AI Analysis Tools A new macOS malware named 'Gaslight' has been identified, attributed to North Korean threat actors. This malware employs prompt injection techniques to confuse AI-assisted malware analysis tools, embedding 38 fake system messages within its Rust binary. The fabricated messages mimic legitimate developer logs and error reports, aiming to mislead AI systems into aborting their analysis. SentinelLabs reported that the malware not only steals data but also provides remote access to attackers. It uses Telegram's Bot API for command and control, ensuring encrypted communication. The malware's unique approach targets the analysis tools rather than traditional sandbox evasion methods, marking a significant evolution in malware tactics. Apple’s XProtect has flagged the malware, indicating its detection capabilities. The malware is currently active and poses a significant threat to macOS users. 13 sources 73

Extracted Entities

New Gaslight macOS Malware Uses Prompt Injection to Disrupt AI

The full story

Extracted Entities

Attack Types (1)

Malware (1)

Platforms (1)