The move comes as regulators increase scrutiny of chatbots and their impact on young people’s mental health

However, all three chatbots inconsistent in answering intermediate-risk questions about suicide