ChatGPT’s safety guardrails allegedly loosened — because clicks matter more than care

A bereaved family suggests OpenAI deliberately weakened ChatGPT’s self-harm prevention safety guardrails to drive more user engagement.