WhatsApp to Fight Scammers with AI-Powered Reminder Feature

Technology company Meta is preparing a new feature to combat digital fraudsters on its WhatsApp service, called Scam Alert. The feature is currently in beta testing and utilises artificial intelligence (AI) to flag potential scam messages by running a machine learning model directly on the user’s device. Meta assures that the data collected by the AI will not compromise user privacy, and no conversations will be shared with WhatsApp, Meta, or any other entity. To use the feature, users will need to activate Scam Alert in the WhatsApp settings menu. Once enabled, an AI model will be automatically downloaded to the user’s phone to learn and identify scam messages; users can deactivate the feature at any time. Scam Alert operates in the background, classifying incoming text messages from unknown senders. The application analyses communication patterns associated with scams, such as requests for personal information, urgent tones, and deliberate trust-building tactics used in romance scams. Upon receiving a notification, users can choose to block the sender, report the conversation, ignore the notification, or mark the unknown sender as trusted. Selecting the trust option removes the warning and stops future notifications for that specific conversation. As the AI continuously learns, users may be directed to optionally share the last five messages from trusted contacts with WhatsApp to help improve the model’s accuracy in distinguishing between scam messages and everyday communication. This data sharing is not automatic and requires the user’s explicit consent. The Scam Alert feature is currently in a limited beta version, so it is not yet available to all users, and Meta has not announced a schedule for a full public launch.