AI brief
A LessWrong post argues that a language model asked about its own personality as the first message of a conversation has no in-context data about its behavior and must rely on cached, zero-shot knowledge.
Written by AI from LessWrong's published text.
Source
“Imagine you ask a language model "What is your personality?" as the first message of a conversation, with no system prompt. The model has no data about its own behavior in its context, so its answer can only rely on cached, zero-shot knowledge.”
Interpretation or community commentary rather than straight reporting.
Read original story ↗