WhatsApp has begun rolling out an optional "Scam Alert" feature that uses machine learning to detect and warn users about potential scam messages targeting them.
The new feature leverages a local machine learning model to identify suspicious messaging patterns commonly associated with scams. Users who enable Scam Alert will receive notifications when the system detects incoming messages matching known scam behaviors.
The machine learning model runs locally on users' devices, meaning scam detection occurs without sending message content to WhatsApp's servers. This approach maintains the end-to-end encryption WhatsApp users expect while adding an additional security layer.
The rollout is optional, allowing users to decide whether to activate the feature. WhatsApp has not specified a timeline for full global availability, but the gradual deployment suggests the company is monitoring performance and user feedback during the initial phase.
This addition addresses growing concerns about scam messages on messaging platforms. Users can report suspicious messages through WhatsApp's existing reporting mechanisms, contributing to the platform's ongoing efforts to combat fraud and abuse.
Hundreds of Customs and Border Protection employees allegedly exploited government databases to conduct unauthorized lookups on romantic interests and colleagues, according to records obtained by WIRED.
Taiwan's Ministry of Digital Affairs detected AI-assisted cyber-attacks targeting government agencies beginning July 20, marking what officials describe as a new threat category. The attacks reportedly originated overseas and have been characterized as a first-of-a-kind breach.
A data theft campaign exploits misconfigured Salesforce Experience Cloud and ServiceNow customer portals to harvest sensitive information. The attackers use custom tools to access data exposed to anonymous users.
A compromised AI package exposed credentials from 2,500 users in a large-scale supply-chain attack. Attackers scraped and exfiltrated terabytes of sensitive data.