Mumsnet denies using AI for content after prompt appears on message board

A prompt titled AIBU Response Schema v.1.0 unexpectedly appeared on a popular parenting forum in response to a user seeking advice about a family dilemma involving a neurodivergent uncle and a sick child. The prompt contained detailed instructions for an AI to write an “interpersonal reasonableness” reply, including definitions of forum shorthand (AIBU, ND, DD), constraints such as “do not diagnose anyone” and “prioritize compassion, boundaries, and de-escalation,” and a request for a clear forum-style ruling with brief actionable advice. Long-standing users called the presence of the prompt “creepy” and “disappointing,” copied it into AI tools and confirmed it could generate a full reply, and interpreted the incident as evidence that machine-generated content might be seeping into conversations traditionally written by forum members.

The forum’s management initially offered conflicting explanations - saying a title-suggestion tool that uses an OpenAI model was responsible, then suggesting the prompt came from a user’s device - before the founder said the system that sends draft titles to OpenAI had apparently returned a full-reply prompt, which she blamed on an external error. Management insists it does not use AI to author threads or replies, while acknowledging use of AI for coding, ads, analytics and moderation flags. Users remain unconvinced, amid broader concerns highlighted by research that AI assistance can produce “semantic garbage” and feel robotic, and a membership decline that has increased sensitivity to preserving human-authored advice.