Technology term
Model Refusal
A response in which an AI model declines to fulfill a user's request, usually because a safety rule, policy constraint, uncertainty threshold, or system instruction has been triggered. Refusals may be appropriate safeguards or, when overly broad, reduce usefulness.
English
Model Refusal
Arabic
رفض النموذج للإجابة
First appeared in
Your AI May Be Carrying Censorship Across Borders Without Telling You