Unverified50% confidenceFactExact time
研究人员将一种在恶意VBScript中嵌入核武器制造说明提示词、以触发LLM安全拒绝机制从而停止分析恶意逻辑的技术称为GuardBreaker
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
GuardBreaker攻击：用核武器提示词瘫痪AI恶意软件分析
Related Entities
Cite This Claim
Stable URI
https://kongchang.com/claim/1000238API
curl https://kongchang.com/api/v1/knowledge/claims/1000238MCP
get_claim(id=1000238)