Unverified50% confidenceFactExact time
Indirect prompt injection (IPI) is an attack that hides malicious instructions within external data that the model is asked to process, as distinct from direct injection where the user inputs malicious instructions directly.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedPrompt injection can be divided into two types: direct injection and indirect injection; indirect injection is where an attacker hides malicious instructions in external data sources such as web pages, PDFs, and emails, and when an AI agent processes this data the malicious instructions are executed as legitimate instructions.78% similarUnverifiedIndirect Prompt Injection attacks embed instructions in external data processed by the AI, and since 2023 security researchers have demonstrated this type of attack in multiple AI products.72% similarUnverified多用户场景下「提示注入（prompt injection）」攻击面会显著扩大，对AI系统的安全防护机制提出更严格要求69% similarUnverified研究人员已发现针对AI代码建议的提示注入、通过恶意仓库影响模型输出的数据投毒，以及利用Copilot上下文窗口实施的信息泄露攻击等新型攻击路径69% similarVerifiedThe attack exploits the automatic execution characteristics of AI agents to conceal malicious behavior within normal operational workflows, bypassing static code scanning tools.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/998451API
curl https://kongchang.com/api/v1/knowledge/claims/998451MCP
get_claim(id=998451)