Google launches AI-powered scam detection feature for Android users
Online fraud is becoming an obsession for many smartphone users. In response to this reality, Google has recently launched a promising new feature that uses artificial intelligence (AI) to help Android users detect and prevent fraudulent behavior.
With Android devices constantly being targeted by scammers via calls and messages, Google has deployed advanced AI technology to protect users.
This feature is designed to automatically detect and warn about calls and messages that show signs of fraud, helping users avoid risks such as losing money, having personal information stolen, or being attacked by malware.
Why is Google launching a new scam detection feature for Android users?
In order to protect users from increasingly sophisticated and damaging phishing attacks in recent times, Google has launched a new feature to detect and prevent an increasingly popular form of fraud, which is voice call fraud.
In an official announcement on March 4, a Google representative confirmed that new AI-powered scam detection features are now rolling out to Android smartphone users.
Traditionally, anti-spam and anti-phishing protections have focused on blocking before a conversation even starts, but this makes them less effective against sophisticated social engineering scams.
To fix that, Google said, "We've invested in new intelligent AI models that can detect suspicious behavior and provide real-time alerts during conversations while still protecting user privacy."
How does Google's new scam detection feature work?
Google is leveraging AI to improve scam detection in Google Messages, helping protect users even after they receive the first message.
According to Google, AI will work in real-time to identify anomalies in short message service (SMS), multimedia message service (MMS), or rich communication service (RCS).
If suspicious signs are detected, the system will alert users to the potential for fraud, and provide the option to ignore, report, or block the sender.
As for the call protection feature, it's a step up from the initial rollout to Pixel phone users in the US late last year. It uses an on-device AI model to analyze conversations in real time.
"If a caller asks you to pay with a gift card to complete a delivery, AI will immediately alert you via audio notification, vibration, and on-screen warning that the call may be fraudulent," Google explains.
As for rollout timing, AI-powered message protection will first launch in the US, UK, and Canada, before expanding to other countries. Meanwhile, call protection, which uses Gemini Nano technology, will be available to all English-speaking Pixel 9 users in the US.
Mike Britton - Chief Information Officer at Abnormal Security (USA) emphasized that, as non-technical attacks become more sophisticated, especially with the rise of AI-powered scams, AI-based security is no longer an option, but a must.
However, he also warned that cybercriminals are using AI to bypass traditional detection systems by creating highly personalized attacks.
Therefore, AI-based defenses need to go beyond simple keyword recognition, and analyze even the smallest behavioral anomalies to detect malicious intent.
Britton concludes that this approach has proven effective in the business-to-business (B2B) space, where AI helps detect sophisticated email compromises and financial fraud.
He said that applying similar technology to protect consumers is an encouraging step forward, helping to prevent fraud before it causes serious consequences.
Does this new scam detection feature ensure user privacy?
One question that many people are wondering about is how secure these new fraud detection features are, especially since they involve AI analysis and some features are enabled by default. Is privacy really guaranteed? And more importantly, is it necessary to disable these features?
The main issue with the way Google designed this scam detection feature is that it works entirely on the device. In the case of Google Messages scam detection, it only applies to messages from people who aren't in your contacts.
Google claims that message processing happens 100% on the device, protecting user privacy to the fullest extent possible. If you report a conversation to help reduce spam, only information about the sender and a few recent messages will be shared with Google and your carrier, not the entire content of your messages.
Similarly, call protection is also built with privacy in mind. Google explains that call audio is only processed temporarily, and nothing is stored, recorded, or sent to Google or third parties. The only difference is that this feature is turned off by default.
Regarding this issue, Mr. Eugene Liderman - Director of Security & Privacy Products at Google, and Chairman of the Security and Anti-Fraud Group of the World Mobile Association (GSMA) shared: "We are excited to bring users new protections, powered by AI, only available on Android".
“These features are designed with privacy in mind to help protect users against the growing wave of sophisticated scams. Fraud detection uses the power of on-device AI to provide real-time alerts while keeping conversations completely private. Most importantly, users are always in control, they can turn this feature off at any time in their settings,” added Eugene Liderman.
As you can see, Google has set a clear line between AI-based security and personal privacy, keeping you safe from scams and not having to worry about your personal data being collected.