Meta Adds AI-Powered Anti-Scam Warnings Across WhatsApp, Facebook, and Messenger

Meta announced new anti-scam tools on March 11, 2026 for WhatsApp, Facebook, and Messenger, alongside new investments in AI detection and broader enforcement work with law enforcement and industry partners. The company positioned the update as both a product-safety rollout and a scaling effort against increasingly organized scam networks.

The most concrete product change is a new WhatsApp device-linking warning. Meta said scammers sometimes try to trick people into linking a victim’s WhatsApp account to a criminal device by asking for a phone number, a linking code, or a QR scan under false pretenses. WhatsApp will now surface alerts when behavioral signals suggest a linking request may be suspicious, showing where the request is coming from and warning that it could be a scam.

Meta also said it is adding alerts for suspicious friend requests on Facebook and expanding advertiser verification. The company wants verified advertisers to account for 90% of ad revenue by the end of 2026, up from 70% today, focusing the requirement on higher-risk categories while leaving lower-risk local businesses outside the strictest layer.

The scale numbers in the post are significant. Meta said it removed more than 159 million scam ads in 2025, with 92% taken down before user reports. In India alone, it said more than 12.1 million pieces of ad content were banned for violating fraud, scam, and deceptive-practices policies in 2025, and over 93% were removed proactively.

Beyond ads, Meta said it took down 10.9 million Facebook and Instagram accounts associated with criminal scam centers. It also said a recent operation with global law enforcement agencies led investigators to disable over 150,000 accounts associated with scam-center networks in Southeast Asia, while Thai authorities arrested 21 individuals tied to scam activity.

The release matters because it combines user-facing friction, advertiser identity checks, AI-based detection, and cross-border enforcement in one operating model. For platform operators and security teams, the announcement is a reminder that large consumer apps are moving from pure moderation toward layered fraud-prevention systems that mix product design, verification, and network disruption.