Understanding the Risk: A recent behaviour observed in WhatsApp’s integration of Meta AI highlights a potential safeguarding concern. Even when a group chat is archived and muted, if someone in that group interacts with Meta AI, it can override the user’s notification settings—causing alerts to appear despite being disabled. Meta has confirmed this is “by design,” not a bug

If AI interactions can bypass these boundaries:

• It undermines user control, especially for vulnerable users trying to limit exposure to certain conversations or individuals.
• It may lead to unintentional re-engagement with groups or individuals they’ve deliberately muted—potentially including those involved in bullying or inappropriate content.
• It complicates digital safeguarding efforts, as parents and educators may not be aware that AI-triggered notifications can override settings.

What Parents should know:

• Meta AI is now embedded in WhatsApp and cannot be removed or fully disabled
• Its presence is persistent and visually prominent, and it can be activated by others in a group chat.
• The AI’s functionality is still evolving, but its integration has already prompted scrutiny from EU regulators due to concerns over privacy, transparency, and user autonomy

• Talk to your child about how Meta AI works in WhatsApp and what to expect from its notifications.
• Review group chat settings together, and consider whether certain groups should be exited entirely if they are no longer safe or appropriate.
• Encourage open dialogue about any unexpected messages or alerts they receive, especially if they feel uncomfortable or confused by them.
• Stay informed about app updates and AI features that may affect how your child interacts with technology.