Meta is adding to WhatsApp’s security with a new AI-powered feature that will scan for suspicious conversations and alert users to possible scams. The tool is undergoing a trial with a small group of users before being released more broadly.

 

How WhatsApp’s Scam Alert Works


The new feature uses on-device machine learning to look for patterns typically associated with fraud in conversations. If it detects possible scam behaviour, WhatsApp will show a warning inside the chat. The alert is only visible to the recipient, so that users can spot potential risks without exposing their conversations. 

What Users Can Do After a Warning

When an alert is displayed, users can decide how to proceed. As always, they can report the conversation or block the sender, or just keep chatting if they think it’s legit. WhatsApp also allows users to mark a conversation as trusted if they think the warning was triggered erroneously. Users can reportedly turn off the feature at any time. 

WhatsApp Expands AI-Based Security

Scam Alert is a new layer to WhatsApp’s existing security measures. Earlier this year, the platform introduced an AI-powered system for identifying suspicious device-linking requests, which scammers might use to take over accounts. The newest tool is currently in a limited beta test and will be released internationally at a later stage.

 

FAQs

Is WhatsApp’s Scam Alert available to everyone?
No. It is currently being tested with a limited group of users.

Can users ignore a scam warning?
Yes. Users can choose to continue the conversation if they believe it is safe.

Can the feature be turned off?
Yes, Meta says users can disable Scam Alert.

Conclusion

WhatsApp’s new Scam Alert feature will add an extra layer of protection to users against scam conversations. Meta is combining scam detection and privacy by leveraging AI on the device and empowering users to control how they respond to warnings.