🛡 New Gaslight Malware for macOS Targets AI Analysts

A Rust-based malware family named Gaslight has been discovered that utilizes prompt injection techniques. It embeds false system messages into binary files to deceive AI agents and force them into issuing incorrect verdicts.

🌍 The attack targets the 'perception' of the analyst (whether human or LLM). Security tool developers will need to account for the risk of context manipulation through code.

👤 If you use AI for code analysis, remember: a file may be designed specifically to confuse your chatbot.

Source 1: https://www.bleepingcomputer.com/news/security/new-macos-malware-embeds-fake-errors-to-confuse-ai-analysis-tools/