Unverified50% confidenceFactExact time
研究人员Kai Greshake等人在2023年发表的论文《Not What You've Signed Up For》系统性地揭示了间接提示注入的威胁模型
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedIn 2023, Kai Greshake and colleagues at Ruhr University Bochum in Germany were the first to systematically classify and assess the hazards of indirect prompt injection attacks in a research paper.70% similarUnverified2023年研究人员Simon Willison系统性地揭示了间接提示注入的危害，指出模型具备联网搜索或文件读取能力时攻击面急剧扩大59% similarUnverified2023年以来研究人员在AutoGPT、LangChain等主流Agent框架中演示了提示注入攻击的可行性59% similarUnverified2023年卡内基梅隆大学与德国CISPA亥姆霍兹信息安全中心的研究团队发表论文，证明当时主流LLM Agent对间接提示注入攻击高度脆弱，成功率超过80%57% similarUnverified2023年研究人员Riley Goodside首次系统性演示了间接提示词注入的危害57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/535981API
curl https://kongchang.com/api/v1/knowledge/claims/535981MCP
get_claim(id=535981)