WhatsApp Tests On-Device AI Scam Alert Feature

Meta has introduced an optional on-device artificial intelligence feature for WhatsApp designed to detect and warn users about potential scam messages, a move aimed at combating rising cyber fraud without compromising the platform's end-to-end encryption. The feature, called Scam Alert, is currently in a limited beta testing phase and uses a machine learning model that runs locally on a user's device to analyze incoming messages from unknown senders for patterns associated with scams.

The system operates entirely on the user's phone, meaning no message content is sent to WhatsApp, Meta, or any third party for classification. This approach is intended to preserve privacy while providing an additional layer of security. When the on-device model identifies a message as a likely scam, it displays a warning banner in the chat that is visible only to the recipient. The user can then choose to block the sender, report the conversation, ignore the warning, or mark the chat as trusted, which removes the warning and prevents future alerts for that conversation.

WhatsApp states the model is trained on patterns observed in scam conversations that users have previously reported. It analyzes...