A new security feature for Android has been announced by Google. The update targets mobile scams that include fraudulent messages, spoofed telephone calls, malicious links, and the attempted theft of sensitive data. The updates include enhanced scam detection using artificial intelligence, additional security measures for screen sharing, and controls to prevent unauthorized call forwarding. 

It is estimated that over 5 million suspected scam calls and text messages are flagged every month by Google's existing scam detection tools in Google Phone and Google Messages. As a result of the latest enhancements, more sophisticated fraud attempts will be identified, particularly those that use prolonged conversations, fake banking communications, and social engineering to trick users into divulging sensitive information.

There is also an optional setting for Enhanced Protection Mode in Google Messages that is anticipated to be available in the next few weeks. Using Google's cloud-based Gemini models to examine suspicious conversations with unknown senders, this feature strengthens existing on-device detection. By using this approach, deceptive messages can be identified whose fraudulent intent may not be immediately apparent. 

This cloud-based analysis is enabled by Google's Private AI Compute environment, which allows data to be encrypted and processed in a dedicated environment that prevents unauthorized access, including Google's own. As part of the company's efforts to enhance the user experience, artificial intelligence is also being used to examine unfamiliar links before users open them. 

When the Gemini Nano model on the supported device is enabled with Enhanced Protection, a real-time web signal can be used to determine if a link may be unsafe, even for newly created websites not yet included in lists of known malicious domains. In addition to providing additional context, the redesigned warnings will provide further context to users before they open links received from external contacts. 

The Google company is also partnering with Android device manufacturers to expand scam alerts beyond Pixel phones in chat notifications. This feature allows potential threats to be flagged before a messaging application is opened by identifying suspicious messages directly in the notification panel. 

As part of its efforts to combat phone-based fraud, Google is introducing new messaging protections. Android will automatically conceal password fields and payment information during live screen sharing sessions using Chrome and Android WebView. 

The expansion is designed to prevent scammers from viewing sensitive information when requesting victims to share their screens with them. A further safeguard is targeted at call-forwarding abuse, a method that scammers use to redirect incoming calls and intercept one-time passwords or other account verification codes. This feature is currently available on Android 15 and later devices. 

In addition to preventing applications from secretly activating call forwarding in the background, Android will alert users if they are tricked into entering a forwarding code through their phone dialer. Pixel and selected devices will be protected starting in early December. In addition to improving protection against bank impersonation scams, Google is enhancing the verification feature for financial calls. 

Android will verify incoming calls against banks' official applications to determine if a bank has actually initiated them. If the verification is unsuccessful, the call will automatically be terminated and an alert will be sent to the recipient. A recent report by Google indicates that more than 3 million spoofed calls impersonating Banco do Brasil and Ita Unibanco have been blocked since the launch of the feature in June.

The company is collaborating with additional financial institutions in order to expand the protection to more customers. This new measure will be introduced in phases, with availability dependent on the device, the operating system version and the features that are supported. 

Even though these updates add additional barriers against fraudulent calls and messages, their effectiveness is subject to the availability of compatible hardware and the level of security enabled on each device.