Meta tests new AI tool to detect scams on WhatsApp

‘Scam Alert’ feature alerts users to potentially fraudulent chats and offers quick actions

Last updated: August 16, 2026 | 11:23

1 MIN READ

Users can block, report or continue chats when the new ‘Scam Alert’ feature detects suspicious activity.

Meta has begun testing an AI-powered feature on WhatsApp designed to warn users when a conversation may be linked to a scam.

The optional ‘Scam Alert’ feature is being rolled out to a limited number of users, according to The Verge. It displays a warning banner when WhatsApp’s AI detects signs that an active chat could be part of a fraudulent scheme.

Users receiving an alert can choose to block the contact, report the account or continue the conversation. If the warning is triggered by mistake, they can mark the chat as trusted and dismiss the alert for that conversation.

WhatsApp warning: AI agents can spam all your contacts — here’s how

WhatsApp also allows users to share the last five messages from a flagged chat to help improve the feature’s accuracy.

The test builds on Meta’s earlier efforts to tackle scams on WhatsApp, including an automated system introduced this year to identify suspicious requests involving linked devices.

Balaram Menon

Balaram brings more than two decades of experience in the media industry, combining sharp editorial judgment with a deep understanding of digital news dynamics. Since 2004, he has been a core member of the gulfnews.com digital team, playing a key role in shaping its identity. Passionate about current affairs, politics, cricket, and entertainment, Balaram thrives on stories that spark conversation. His strength lies in adapting to the fast-changing news landscape and curating compelling content that resonates with readers.

Get the latest news and updates straight to your inbox