WhatsApp removes 6.8 million fraudulent accounts using AI-driven scams
WhatsApp removed 6.8 million fraudulent accounts in the first half of 2025 that used AI to generate convincing scam messages. Criminals created contacts on dating apps and SMS, then lured victims to WhatsApp to steal money, banking information or identities through schemes involving fake packages, cryptocurrency investments and government impersonation.
What happened
WhatsApp, owned by Meta, removed 6.8 million fraudulent accounts during the first half of 2025, according to a report by cybersecurity firm Kaspersky.
Scammers used artificial intelligence to create well-crafted and convincing messages, often initiating contact on dating apps, SMS or other social media before directing victims to WhatsApp. The fraudulent schemes included false package delivery offers requesting payment, cryptocurrency investment promotions, online game invitations, and government impersonation messages threatening fines.
Reported impact
Kaspersky reported that the automation of AI-driven attacks, combined with the trust users place in WhatsApp, facilitates the spread of scams across Latin America, where most fraud occurs through direct messages to avoid account suspension.
Index analysis
Kaspersky stated that platforms' verification systems remain weak, allowing malicious accounts to be authenticated with only a payment. The firm noted that announced actions by Meta could have limited impact in Latin America, where scammers avoid mass messaging to prevent account suspension.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.