Scholarlink

<p>AI chatbots built on large language models (LLMs) are poised to make a significant impact across health sectors. Prior to widespread implementation of AI chatbots in youth mental health care, it is critical to adopt responsible design and evaluation practices. This is exemplified by the need to account for youth-centric language and youth expressions of mental health to enhance chatbot acceptability and effectiveness and facilitate their safe translation into clinical practice. When designing an AI chatbot for youth mental health it is critical to consider how the LLM has been trained and fine-tuned, ground the chatbot in a curated knowledge base, align the LLM using human feedback, and establish preprogramed guardrails and ethical filters to promote safety and reduce risk of harm. It is vital that AI-chatbots built on LLMs that are intended for use with young people reflect their voices, experiences, needs, and perspectives and adhere to ethical principles common to international AI ethics frameworks to promote fairness, accountability, and safety and to prevent harm.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The implications of AI chatbots built on large language models for youth mental health

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月23日

查看全文
  • Haley M. LaMonica,
  • Ian B. Hickie,
  • Adam Poulsen,
  • Yun J. C. Song,
  • Zsofi de Haan,
  • William Capon,
  • Ashlee Turner,
  • Carla Gorban,
  • Elizabeth Scott,
  • Frank Iorfino

摘要

AI chatbots built on large language models (LLMs) are poised to make a significant impact across health sectors. Prior to widespread implementation of AI chatbots in youth mental health care, it is critical to adopt responsible design and evaluation practices. This is exemplified by the need to account for youth-centric language and youth expressions of mental health to enhance chatbot acceptability and effectiveness and facilitate their safe translation into clinical practice. When designing an AI chatbot for youth mental health it is critical to consider how the LLM has been trained and fine-tuned, ground the chatbot in a curated knowledge base, align the LLM using human feedback, and establish preprogramed guardrails and ethical filters to promote safety and reduce risk of harm. It is vital that AI-chatbots built on LLMs that are intended for use with young people reflect their voices, experiences, needs, and perspectives and adhere to ethical principles common to international AI ethics frameworks to promote fairness, accountability, and safety and to prevent harm.