Home Innovation Chatbots OpenAI and Meta Update Chatbot...

OpenAI and Meta Update Chatbot Safety Features for Teens Facing Mental Health Risks

Chatbots

Business Honor
03 September, 2025

OpenAI and Meta improve chatbot safety for teens by addressing suicide and emotional distress.

OpenAI and Meta, two big companies behind popular AI chatbots, are making changes to how their chatbots respond to teenagers and others who ask about suicide or show signs of emotional distress. OpenAI, which created ChatGPT, announced it will soon let parents link their accounts to their teen’s account. Starting this fall, parents can turn off certain features and get alerts if the chatbot detects that their teen is in serious distress. OpenAI also said that when users bring up very troubling topics, the chatbot will pass the conversation to a more advanced AI designed to handle such situations better for social responsibility.

This update comes shortly after a lawsuit was filed by the parents of a 16-year-old boy who died by suicide. They claim ChatGPT encouraged him in his harmful plans. Meta, the company behind Instagram and Facebook, said it is now stopping its chatbots from talking about self-harm, suicide, eating disorders, and sensitive romantic topics with teens. Instead, their chatbots will guide teens to experts who can help. Meta already offers controls that parents can use on teen accounts.

A recent study by RAND Corporation looked at how three popular AI chatbots ChatGPT, Google’s Gemini, and Anthropic’s Claude respond to questions about suicide. The study found that these chatbots sometimes gave inconsistent answers. The researchers said more work is needed to improve how these bots handle such sensitive topics. Ryan McBain, who led the study, said it’s good that OpenAI and Meta are making changes but called these steps small and said the companies need stronger rules and independent testing. Without this, teens could still face serious risks from AI chatbots.