Archive — US — techcrunch
Meta updates chatbot rules to avoid inappropriate topics with teen users
Meta is updating its AI chatbot policies to protect teen users from inappropriate content following criticism over its previous guidelines that allowed discussions on sensitive topics like self-harm and romantic relationships. The interim changes will limit teens' access to certain chatbots and emphasize safety, with plans for more comprehensive updates in the future.
Key points
- Meta changes AI chatbot training to enhance teen safety after an investigative report revealed past gaps.
- Chatbots will no longer engage with teens on sensitive topics such as self-harm, suicide, and disordered eating.
- Meta acknowledges previous mistakes in allowing chatbots to discuss inappropriate topics with minors.
- New restrictions will be placed on the types of AI characters accessible to teens, limiting exposure to educational content.
- Sen. Josh Hawley initiated a probe into Meta's AI policies following the negative publicity.
- A coalition of 44 state attorneys general expressed outrage over the risks posed to children's emotional well-being.
Why it matters
- The changes come amid growing scrutiny of social media companies and their responsibility to child safety.
- Historical concerns about online platforms' impact on young people's mental health have prompted legislative action.
- Disrupt 2025 reflects ongoing discussions in tech about the ethical implications of AI and its role in society.
Meta's immediate chatbot policy changes underscore the increasing demand for safer digital environments for minors, reflecting broader societal concerns and potential regulatory actions in the tech industry.