Verified65% confidenceFactExact time
Large language models cannot structurally distinguish between user instructions and instructions embedded in content, because both are part of the same token sequence from the model's perspective.
3
Sources
65%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLLMs cannot distinguish instructions from data at a structural level, because both exist in natural language form within the same context window.76% similarUnverifiedLarge language models do not execute code themselves; they are only responsible for calling tools, while the application-layer code actually executes the functions.73% similarUnverified在大语言模型底层，系统提示词、用户输入和模型输出都是同一个token序列的不同片段，依靠特殊分隔符和训练时植入的行为模式维持边界，而非硬件或操作系统级别的隔离73% similarUnverifiedA language model itself can only generate text and cannot directly access a database or know a user's order status; it must use a tool-calling mechanism to perform actual operations.73% similarVerifiedThe root cause of Prompt Injection is that large models cannot structurally distinguish between "system instructions" and "user data," as both are presented as natural language text at input time.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/948685API
curl https://kongchang.com/api/v1/knowledge/claims/948685MCP
get_claim(id=948685)