WhatsApp is testing an optional Android beta feature that warns about suspicious messages from unknown contacts while keeping analysis on-device.
WhatsApp has started limited Android beta testing of an optional Scam Alert feature that warns users about suspicious messages from unknown contacts while keeping message analysis on the phone. The feature is not part of a public rollout yet, but it is appearing for a small group of testers on WhatsApp beta for Android version 2.26.34.1. For users who handle sales, support, or operations chats on WhatsApp, the appeal is straightforward: an extra warning layer on unfamiliar inbound messages without sending chat content to Meta for server-side scanning.
Android beta users start seeing Scam Alert on unknown chats
Scam Alert is built for messages that come from unknown contacts, not for every conversation in the app. It is also optional rather than automatic, which means users can turn it on or off in WhatsApp settings if they are among the testers who have access. When the feature is triggered, WhatsApp shows a warning to the user instead of blocking the sender or deleting the message on its own. That keeps the tool in the category of decision support, not a full anti-scam filter.
The current rollout is narrow. WhatsApp beta version 2.26.34.1 is the concrete build tied to the feature, and availability is still limited to a small Android beta group. That matters for teams trying to plan around it: this is something to watch in testing, not a protection you can count on across an entire business account or staff fleet today.
How WhatsApp handles the detection on-device
The most notable part of the feature is where the detection happens. WhatsApp is using an on-device machine-learning model to classify suspicious messages locally on the phone, which means the message content does not need to be uploaded to Meta for analysis as part of this feature. Meta’s position is that this preserves WhatsApp’s end-to-end encryption model because flagged content is not automatically shared with WhatsApp or Meta.
That design shapes the user flow after a warning appears. A flagged chat can still be opened, and the user decides what to do next: block the sender, report the chat, or continue the conversation. In one account of the beta behavior, users can also mark a false positive as trusted so the warning stops appearing for that chat. That last option is still best treated as beta-specific, but the broader point is clear: the feature adds a warning step without taking control of the conversation away from the user.
What’s still missing from the rollout
The biggest open question is timing. WhatsApp has not announced a public release date for Scam Alert, and there is still no confirmed iPhone rollout or wider platform plan. The exact settings path for enabling the feature is also not consistently documented, so early testers may not all see it in the same place.
The beta group remains small, which leaves broader tester access unclear for now. For business users, that means Scam Alert is best understood as an early security feature in development rather than a new default safeguard inside WhatsApp. If WhatsApp expands the test, the details worth watching are simple: whether more Android users get it beyond version 2.26.34.1, whether the trusted-chat control stays in place, and when WhatsApp says the feature is ready for a public release.
Disclaimer: This article was created with the assistance of AI. Images are for illustrative purposes only.
About the author
Samarth Agrawal is an AI and technology professional who writes about WhatsApp, automation, and emerging AI trends. He focuses on simplifying complex tech updates into practical insights for businesses, creators, and everyday users
