Unverified50% confidenceFactExact time
LLMs are architecturally incapable of distinguishing trusted instructions from untrusted input, because they process system instructions and user input as the same text sequence, and a purely technical solution analogous to parameterized queries has yet to emerge.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLLMs cannot distinguish instructions from data at a structural level, because both exist in natural language form within the same context window.78% similarVerifiedThe root cause of prompt injection lies in the fact that LLM architecture encodes system prompts, user inputs, and externally retrieved content uniformly into the same continuous token sequence, making it impossible for the model to distinguish trusted instructions from untrusted data at the semantic level.76% similarUnverified提示注入攻击的根源在于LLM在推理时无法从语义层面可靠区分「系统指令」与「用户数据」，两者均以自然语言形式存在于同一上下文窗口中75% similarUnverifiedLLM是强大的语言理解与生成工具，但不应被默认当作确定性的计算与执行系统74% similarUnverifiedLLM处理的是token（词元）而非单个字符，导致在字符计数任务上存在天然缺陷73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829746API
curl https://kongchang.com/api/v1/knowledge/claims/829746MCP
get_claim(id=829746)