A new wave of phone scams powered by increasingly convincing AI-generated voices is making it harder than ever to tell what’s real — and what’s not —but a major shift in mobile security may soon change that. Google is rolling out a new fake call detection capability on Android designed to spot AI voice scams in real time, flagging suspicious calls before users can be manipulated by impersonations that sound alarmingly authentic. As cybercriminals refine their use of voice cloning to pose as trusted contacts, banks, or even government officials, this feature could become a critical line of defense. Keep reading to learn more about how Google is rewriting the rules of mobile defense.
Google to launch new mobile defense tool
Imagine answering your phone and hearing a voice that sounds exactly like your mom, your boss, or your bank. The caller’s voice has the right tone, and your phone even displays their name and number from your contact list.
But there’s a catch: the person on the line is actually a scammer using artificial intelligence (AI) to clone their voice and fake their phone number.
This type of fraud is skyrocketing. According to a global report by INTERPOL, these high-tech impersonation scams have tricked people out of a staggering $400 billion worldwide. To fight back, Google is rolling out a brand-new Fake Call Detection feature for Android phones to catch these tricksters in the act.
Why traditional spam filters fall short against AI voice scams
In the past, scammers used random numbers, making them easy for your phone to flag as spam. Today, they use advanced software to do two things:
- Number spoofing: They manipulate the network to make it look like a trusted friend or family member is calling you.
- Voice cloning: By copying just a few seconds of audio from a social media video, criminals can make a computer perfectly mimic someone you love.
Because these calls look and sound real, they trick people into panic-buying gift cards, wiring money, or giving away passwords during fake emergencies.
How Google’s Fake Call Detection works
Google’s Fake Call Detection doesn’t just look at who is calling — it double-checks if the call is actually real. The new feature tackles both phone number spoofing and voice manipulation simultaneously through an automated verification process. Operating silently in the background, the technology acts as a cryptographic handshake:
- The signal: When a saved contact calls your device, their phone automatically sends a silent, encrypted confirmation token via Rich Communication Services (RCS) protocols.
- The verification: If an imposter spoofs that number from an unauthorized external server, your phone will recognize that this crucial confirmation signal is missing.
- The background check: Sensing an anomaly, your phone instantly executes a silent background ping to the actual contact’s physical device to ask if they are currently making a call.
- The alert: If the real device confirms it is inactive, an immediate warning pops up on your screen advising you that the caller is an imposter and that you should hang up right away.
Will my privacy be protected?
A major concern with any call-screening tool is privacy. Google designed this system to run entirely on your physical device. The cryptographic handshake remains end-to-end encrypted through the open RCS framework, meaning no voice data, audio recordings, or call transcriptions are stored or transmitted to external Google servers.
Who gets the new feature?
The feature is rolling out to smartphones running Android 12 or newer that use the standard Phone by Google app as their main dialer. It is turned on by default, meaning you don’t have to navigate confusing settings menus to stay safe.
And because the framework relies on open RCS standards, Google has positioned this tech as an industry blueprint, allowing other smartphone manufacturers and alternative dialer apps to adopt the protocol and collectively fortify the global mobile ecosystem against AI-driven threats.
Source: Cybersecurity Insiders