Hackers Can Hide Malicious AI Commands Inside Normal English to Bypass Security Filters
Check Point's PuzzleMask technique hides malicious prompts in ordinary English that fast gatekeeper models miss but high-reasoning downstream models execute.
Check Point researchers disclosed PuzzleMask, a technique concealing policy-breaking instructions in natural-language prose without encodings or invisible characters. Fast screening models classified all 23 crafted wrappers as safe, while a high-reasoning model recovered and acted on the hidden instruction in 17 of 18 tests (94.4%). The gap stems from capability imbalance between gatekeeper and target models, with defenses including paraphrasing untrusted input, stricter self-referential wording rules, and output/tool-call monitoring.
- 23/23 crafted prompts bypassed lightweight gatekeeper models
- High-reasoning model executed hidden instruction in 94.4% of tests
- No Base64, emoji, or invisible-character artifacts used
- Mitigations: paraphrase untrusted content, monitor tool calls, limit agent permissions
- ATT&CK
- T1036 Masquerading
- Vendors
- Check Point
- AI models
- gpt-5-thinking-high
From the original article
4 min read at cybersecuritynews.com
A newly disclosed AI attack technique shows that harmful commands do not need strange symbols, hidden text, or coded strings to evade security checks. Attackers can conceal an instruction inside ordinary English and rely on different AI models to interpret the same text differently. The method, called PuzzleMask, targets applications that put a fast screening […]
By Tushar Subhra Dutta · cybersecuritynews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.