With more than three billion monthly active users worldwide, WhatsApp can be a fertile ground for scams as fraudsters constantly evolve their tactics. To help protect its users, the Meta-owned app has begun testing a new feature that will alert users to potential scam messages and help them identify attempts to manipulate them.
The Scam Alert, Meta said, is an optional feature that runs an on-device machine learning model to flag potential scam WhatsApp messages. The tool is currently rolled out on a limited scale before becoming available to all WhatsApp users.
In late June, WhatsApp said its users will no longer need to share their phone numbers with others to chat with them. Instead, they can pick unique WhatsApp usernames to share, in a move “designed to protect the privacy of your phone number”, it said when announcing the update. The move, however, raised concerns about fraud and sparked a question: could scammers use usernames to impersonate people, brands or public figures? To help alleviate concerns around scammers, the platform is currently developing Scam Alert.
Stay up to date with the latest news. Follow KT on WhatsApp Channels.
What is Scam Alert?
Scam Alert is a new WhatsApp tool that is now being tested. It aims to help alleviate concerns regarding scams on the social media platform. Meta said recent advances in on-device machine learning models make it possible to run accurate text classification entirely on mobile hardware.
How does it work?
Once the user turns it on, Scam Alert downloads a machine learning model to the device, where it runs processes to classify whether incoming messages from non-contacts match known scam patterns. The model is trained on patterns observed in scam conversations from reports that users have sent to Meta. It performs probabilistic classification based on conversational structure and linguistic signals.
“No content is automatically reported to WhatsApp, Meta, or any third party,” according to Meta.
What happens when it detects a scam threat?
If the model identifies a message as a likely scam attempt, the user sees a warning in the chat, which is not visible to the other person. From there, the user can decide what to do: block, report, or continue the conversation.
What if the user marks the message as safe?
If the user decided that a warning is incorrectly flagged, they can mark the chat as trusted, in which case the warning is removed and Scam Alert will not flag that chat again. If a user marks that they trust a chat, they can also opt in to share the last 5 messages received with WhatsApp to help improve the feature’s accuracy.
Will it automatically flag scam alerts?
No, Meta clarified that there will be no automatic reporting. “WhatsApp is unable to initiate sharing of any user data without the user’s action. The only way message content, or even the fact a scam was detected, reaches our servers is if the user explicitly chooses to report it, which is consistent with how user reporting works on WhatsApp.
Can users choose to turn it off?
Yes. Scam Alert is a user-controlled tool that provides additional information, and the user can turn it off or on at any time.
Will the AI model and its data remain on the device?
Yes. The model and the message data it processes will all remain on the device. “All inference happens on-device and no message content leaves the user device for classification,” Meta said in a statement.
When will it be available to all WhatsApp users?
Scam Alert is now being tested on a limited scale in Beta. Meta has not revealed yet when it will be made available for everyone.
Scammers will continue to evolve their tactics. To keep staying ahead of them, so will we. We welcome feedback from users, security researchers, and the broader security community
Meta
Source: Khaleej Times