Unverified50% confidenceFactExact time
依赖混淆攻击由安全研究员Alex Birsan于2021年首次公开披露
1
Sources
50%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
GPT-5.6 Sol Limited Preview: The Safety Boundaries Behind AI's Capability Expansion
bilibili江枫渔火AI7/9/2026
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.68% 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%62% similarUnverified提示注入攻击的概念最早由安全研究员 Riley Goodside 于 2022 年 9 月在 Twitter 上公开演示62% similarUnverifiedThe project was shared on Bilibili by its creator58% similarPartially VerifiedThe content of this article is compiled from hands-on evaluations shared by a Bilibili content creator.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/487542API
curl https://kongchang.com/api/v1/knowledge/claims/487542MCP
get_claim(id=487542)