WhatsApp is working on a new safety feature that shifts scam detection closer to the user’s device, rather than relying on server-side analysis.
The tool, currently in development for Android, is designed to flag suspicious messages from unknown contacts directly within the chat. If the system detects patterns linked to common scams, it will surface a warning banner, giving users a clearer signal before they engage.
WhatsApp plans to process messages entirely on the phone, keeping conversations within the boundaries of its end-to-end encryption model. That means the company doesn’t need to read or store message content externally to identify potential fraud.
Also read, WhatsApp’s New Incognito AI Mode Shows Meta Wants to Lean into Privacy
The feature doesn’t block messages outright. It nudges the user instead. Once a warning appears, you can either trust the sender and continue or block and report the contact. That design choice keeps control with the user while adding a layer of context at the point of interaction.
There’s also a quiet logging system in the works as activity tied to scam alerts will stay on the device, allowing users to review when warnings were triggered without sending that data back to WhatsApp.
This shows a broader shift in how messaging platforms are approaching safety. Scams on WhatsApp often rely on speed and emotion, pushing users to act before thinking.
Moving detection on-device reduces delay and keeps the warning close to the moment of decision. It also avoids the privacy trade-offs that come with scanning messages in the cloud.
The feature is still being tested and isn’t available to users yet. If it rolls out as expected, it could reshape how WhatsApp balances privacy with real-time protection, especially in markets where messaging apps double as financial and business channels.
Don’t miss important articles during the week. Subscribe to techbuild.africa weekly digest for updates.
