Google Messages tests AI scam detection to flag risky texts and links.

Sometimes a message that asks you to pay twenty rupees to release a parcel or to complete a bank KYC update can look very real. Many people might think those texts are alerts from a delivery company or, from a bank. Sometimes those messages are actually scams designed to steal money or personal information.

Google is working on a new way to help users spot such threats before they fall for them. Its Google Messages app is testing “Enhanced Protection", a feature that uses Gemini AI to detect possible scams within conversations and examine potentially dangerous links sent by unknown contacts.

Scam Detection Could Look At The Entire Conversation

Fraudsters rarely reveal Fraudsters intentions immediately. A scammer may first act as a delivery agent, bank employee or customer support representative catching recipients’ eye before shifting conversation. After gaining recipients attention conversation may slowly shift toward requests, for payments KYC updates, OTPs or personal information.

Google's new protection system is designed to look for these patterns across an entire conversation rather than evaluating just one message. This could make it easier to identify scams that develop gradually or use less obvious wording.

Google says it uses cloud-based Gemini models for this task. The company processes the information through Private AI Compute, which it describes as a protected environment where data is encrypted and inaccessible to Google.

Gemini Nano Will Also Check Suspicious Links

Enhanced Protection also introduces link checking on supported devices. Google's Gemini Nano model can analyse unfamiliar links directly on the phone, using information from the web to identify potentially fraudulent websites that may not yet have been added to existing blocklists.

The feature can also provide more prominent warnings when a link comes from someone outside the user's contacts. The goal is to encourage users to pause and verify a link before opening a website that could result in a fraudulent payment or expose personal information.

However, support will vary depending on the device. Link checking is available on the Pixel 10 series and newer, Samsung Galaxy S26 lineup, and Galaxy Z Fold 8 and Z Flip 8, along with some other devices. Conversation-based detection is expected to support the Pixel 6 series and later, as well as the Galaxy S24 series and later.

When Will Enhanced Protection Arrive?

Google has not announced an India launch date for the feature. Enhanced Protection will also be optional, meaning users will need to activate it after it becomes available on their device.

To check whether it is available, open Google Messages, tap your profile picture and select ‘Messages settings’. Go to ‘Protection and safety’, followed by ‘spam and scam protection’. When the feature arrives on your phone you can choose Enhanced Protection of Standard Protection or No Protection.

Until the feature arrives users should handle payment requests and KYC messages, with care. Users should verify those requests through the company’s app or website and users should never share an OTP with anyone who contacts you unexpectedly.

ABOUT THE AUTHOR

Kahekashan

[email protected]

Kahekashan is a passionate technophile with a keen eye for cutting-edge gadgets, emerging technologies, and everything in the digital realm. Raised in a Defence family with strong values and a background in literature, she has consistently pursued excellence in every endeavour. Her last full-time assignment involved content writing with the Indian School of Business.

Read More