Meta launches anti-scam tools on Facebook, WhatsApp and Messenger: Here’s what is new

Meta has announced that it is taking an aggressive step against online scammers with a set of new tools, expanded AI detection systems and fresh law enforcement partnerships. These new tools are designed to protect users across Facebook, WhatsApp and Messenger. The company also says that it has built advanced AI systems capable of analysing multiple signals at once to catch impersonations faster.

Meta’s new tools for scam detection rolling out now

Facebook: Warnings on suspicious friend requestsMeta says that it is testing a new alert system on Facebook that warns users when a friend request shows signs of suspicious activity. For example, when the sender shares very few mutual friends, or when their profile lists a different country than expected. In such cases, users will be alerted to block or reject the request before engaging.WhatsApp: Alerts for suspicious device linkingMeta notes that one of the clever scam tactics is tricking users into unknowingly linking their WhatsApp account to a scammer’s device.

This can happen through fake talent competitions, voting schemes or fraudulent QR codes. The new tools for WhatsApp will now detect when a device linking request looks suspicious and alert the user before the link is established. This shows where the request is coming from and flagging it as a potential scam.Messenger: Advanced scam detection in more sountriesComing to Messenger, Meta is expanding its AI-powered scam detection on the app to more countries this month.

When a conversation with a new contact shows patterns associated with common scams, such as suspicious job offers, users will be warned and asked if they'd like to share recent messages for an AI review. If a scam is detected, Messenger will explain what type of scam it may be and suggest actions including blocking or reporting the account.

Meta AI to take on celebrity impersonation and fake brands

Another persistent and damaging forms of online fraud involves scammers posing as celebrities, public figures or well-known brands and use fake fan pages, misleading bios and other deceptive tactics to lure victims.Meta says it has built advanced AI systems capable of analysing text, images, and surrounding context simultaneously to catch these impersonations faster and at greater scale than traditional detection methods allowed. The announcement comes with a striking set of numbers that illustrate just how large the problem has become. In 2025 alone, Meta says it removed over 159 million scam ads – 92% of which were taken down before anyone even reported them. The company also disabled 10.9 million accounts on Facebook and Instagram linked to organised criminal scam centres.