Unverified50% confidenceFactExact time
Achieving 'separation of data and instructions' through system prompts alone is insufficient to defend against prompt injection; as models grow larger and contexts grow longer, models instead become more inclined to follow instructions in the data.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSystem prompt adherence is not a hard constraint but a probabilistic tendency; models may ignore system instructions when user input creates a semantic conflict with the system prompt.75% similarUnverifiedWhen constraining output format through prompts, the model may forget these constraints in long conversations or complex reasoning scenarios.75% similarUnverified模型在多步推理上越强，当被授予对数据环境的完整访问权限时，其执行路径就越不可预测73% similarUnverifiedTurning 'context retention' from a model capability problem into an engineering problem, feeding the right information at the right time through an orchestration layer, is more controllable and easier to debug in long-horizon tasks.73% similarUnverifiedSmaller models can handle fewer instructions than larger models, and non-reasoning models can handle fewer instructions than reasoning models.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048824API
curl https://kongchang.com/api/v1/knowledge/claims/1048824MCP
get_claim(id=1048824)