WhatsApp has begun a limited beta rollout of a new "Scam Alert" feature designed to warn users about potential scam messages. This optional feature utilizes a local machine learning model to identify and flag suspicious conversations. The company is currently testing the system with researchers participating in its Bug Bounty community.
According to WhatsApp, the Scam Alert feature operates entirely on the user's device. The machine learning model processes incoming messages from non-contacts to detect patterns indicative of known scams, relying on linguistic signals and the conversational structure. Crucially, no message content is transmitted off the device for classification, nor is it automatically reported to WhatsApp, Meta, or any third party, thereby complementing the platform's end-to-end encryption.
When the model identifies a potentially fraudulent message, a warning appears in the chat, offering the user options to block the sender, report the conversation, or continue interacting. Users can also dismiss a warning if they believe it was incorrectly flagged, marking the chat as trusted. In such cases, the warning is removed, and Scam Alert will not flag that specific chat again.
Users who mark a chat as trusted are also given the option to share the last five messages received with WhatsApp. This data is intended to help improve the accuracy of the feature. WhatsApp emphasizes that both the message data processed by the local model and the model itself remain on the device, and users retain the ability to disable the Scam Alert feature at any time.
This initiative is part of a broader effort by WhatsApp to enhance user protection against scams and sophisticated threats. In March, Meta announced that WhatsApp would introduce warnings for device-linking requests that exhibit behavioral signals of fraud. This addresses a common tactic where attackers trick users into scanning malicious QR codes or sharing linking codes to hijack accounts.
Two months prior, WhatsApp also started rolling out "Strict Account Settings." This security feature is aimed at high-risk individuals, such as journalists and public figures, to protect them from advanced threats, including spyware attacks. The introduction of Strict Account Settings followed incidents where various individuals had their devices infected with spyware, including NSO Group's Pegasus, through zero-click exploits in messaging applications.
WhatsApp's messaging service is widely used, with over 3 billion people across more than 180 countries relying on it for communication. The new Scam Alert feature aims to add another layer of defense against the evolving landscape of online fraud.
