New Tool Evaluates AI Chatbots for Mental Health Advice

A new research tool tests how well AI chatbots support teenagers, revealing inconsistent advice quality across major platforms.

Updated on Oct. 30, 2026 in Teens

Isometric editorial illustration of nested geometric cubes with floating organic nodes, representing the evaluation of digital mental health tools.
Researchers have launched a new evaluation tool to measure the accuracy and clinical safety of mental health advice provided by AI chatbots. AI Illustration. Upload story photo >

Live Poll

Do you trust artificial intelligence chatbots to provide accurate mental health advice to teenagers?

Researchers have introduced the Adolescent Mental Health AI Response Tool to measure how well chatbots provide mental health advice. The study tested four popular AI platforms across seven domains to see how they handle clinical guidance for teenagers.

Why it matters

As more adolescents turn to AI for mental health support, this tool helps determine if chatbot responses align with clinical standards. The findings reveal inconsistent advice quality, particularly when comparing anxiety versus depression support.

The new tool evaluated four major AI chatbots, including ChatGPT-5 and Gemini 2.5, across seven distinct mental health domains. Scores showed that these models generally provided better information for anxiety than for depression.

The players

RAND Corporation

A research institution that provides data-driven policy analysis and recently focused on AI safety for youth mental health.

The details

Researchers developed the tool by having clinicians verify candidate items for objective evidence and clinical importance. They then used an ensemble of three large language models to score how chatbots handled multi-turn conversations about mental health. The results indicate that chatbots are currently more capable of addressing moderate clinical presentations than severe cases.

Timeline

  1. October 30, 2026: The research findings were officially published.

The Home Front

This development follows a trend of increasing academic scrutiny regarding the role of generative AI in family life and youth health. It marks a push to establish clinical benchmarks for digital tools that teens often access independently in their bedrooms.

Parents should maintain open conversations with teenagers about the risks of using AI chatbots for medical or psychological advice. It is best to verify any health-related guidance obtained from AI tools with a licensed mental health professional.

The takeaway

AI chatbots currently provide inconsistent mental health guidance for adolescents that does not replace professional care. Always treat chatbot advice as an information supplement rather than a clinical diagnosis or treatment plan.

Further reading

For more on managing digital safety at home, visit the Teens section.

More information

Access the full findings on the RAND research publication page.

Source note: This article includes information reported by Rand Corporation.

Live Poll

Do you trust artificial intelligence chatbots to provide accurate mental health advice to teenagers?