Google Messages is getting an opt-in Enhanced protection mode that uses cloud Gemini models to assess suspicious conversations with unknown senders. It also adds stronger link checks on supported phones. Google announced the update on October 14, with a rollout over the following weeks, but has not given a separate India launch date.
What Enhanced protection mode adds in Google Messages
Scam Detection in Google Messages already flags many traditional and conversational scams. Enhanced protection is a separate opt-in setting that Google says unlocks its strongest defences. Google says it will roll out in the weeks following the October 14 announcement.
With Enhanced protection enabled, Google describes three upgrades:
- Cloud Gemini analysis through Private AI Compute. For chats with an unknown sender, Messages can use Google’s more capable cloud Gemini models to judge the full conversation context and a scammer’s intent faster. Google says this sits on top of existing on-device detection.
- Smarter link safety with on-device Gemini Nano. Messages already flags known bad URLs. Enhanced mode adds real-time web signals so Gemini Nano can judge unfamiliar links before you tap, including brand-new sites that are not yet on blocklists.
- Clearer checkpoints for unfamiliar links. Redesigned warnings give more context and a pause before opening links from numbers that are not in your contacts. Google says this style of warning was previously limited to select countries or devices with Android Advanced Protection.
Google also says AI-powered Scam Detection across Phone by Google and Google Messages already flags more than 5 million suspected scam calls and texts every month.
Privacy: what Private AI Compute means here
The privacy claim matters because Enhanced protection sends conversation context to the cloud for harder cases. Google says that cloud analysis runs in Private AI Compute, where the data is encrypted and processed in a dedicated space that Google describes as private to you and inaccessible to anyone else, including Google.
That is different from fully on-device AICore features, which keep processing local. Enhanced mode deliberately combines both: on-device checks plus secure cloud Gemini when the chat looks complex.
Google also notes you can optionally report a text as spam, which sends the message contents to Google for analysis. That reporting path is separate from the Private AI Compute description above.
Which devices get which pieces
Availability splits across features. The cloud checks, on-device link analysis and other Android safeguards have different device requirements:
| Feature | Device notes from Google |
|---|---|
| Enhanced Scam Detection with Private AI Compute | Pixel 6 and up; Samsung Galaxy S24 series and up; Galaxy Z Flip6 / Z Fold6 and up; other Android AICore devices |
| Gemini Nano link safety checks | Pixel 10 and up; Galaxy S26 series; Galaxy Z Flip8 / Z Fold8; other select devices |
| Unfamiliar-link checkpoints | Android 6 and newer |
| Chat-notification scam alerts (OEM expansion) | Later this year with partners including HONOR, Infinix, iQOO, itel, JOVI, Motorola, OPPO, OnePlus, realme, TECNO, vivo, and Xiaomi; timing varies by maker |
| Screen-share privacy (Chrome / WebView blanking) | Rolling out in October 2026 on Android 15 and newer |
| Call-forwarding protection | Early December 2026 on Pixel and other select devices |
An Android 6 or newer phone meets the stated requirement for unfamiliar-link checkpoints, but that does not mean it can run Gemini Nano link analysis. The stronger AI checks require the hardware listed for that feature, as well as access to the rollout.
Earlier Pixel scam alerts also varied by region and app. Owning a supported phone alone does not guarantee that every protection is available.
Is Enhanced protection coming to India?
Google has not confirmed an India date or a complete India-specific feature list in this announcement. Several manufacturers popular in India are part of the notification-alert expansion, but models and timing will vary.
The notification expansion is separate from Enhanced protection in Messages:
- More manufacturers are included. Several makers popular in India (Motorola, OPPO, OnePlus, realme, vivo, Xiaomi, iQOO and others) are named for later chat-notification scam alerts. Google still says device availability varies by manufacturer.
- Verified bank-call spoof blocking in the same blog is described with Brazilian bank partners and Revolut. Do not assume Indian bank numbers are covered unless Google or your bank later confirm it.
Keep Google Messages and system services updated, and check for the setting as it rolls out. An absent warning does not make an unexpected OTP, KYC or courier-payment message safe.
How to check for Enhanced protection
Enhanced protection is an opt-in setting inside Google Messages. Google has not yet specified its exact menu path in the announcement.
When the setting reaches your phone:
- Update Google Messages from the Play Store and install available system updates.
- Check the app’s settings for Enhanced protection when the rollout reaches your account.
- Read the explanation of cloud processing before enabling it.
- If the option is missing, check Google’s availability information again later. Updating the app does not guarantee immediate access.
Also keep AICore up to date if your phone supports it. Google’s Android Help page lists scam detection among AICore-powered features, and Enhanced protection’s Private AI Compute footnote points to AICore-capable devices. You can find AICore under Settings → your Google profile → All services → Privacy & Security → System services.
Other Android scam safeguards in the same update
Google also announced protections against scammers who ask you to share your screen or redirect incoming calls:
- Automatic screen-share privacy. During a live screen share, login and payment fields in Chrome and Android WebView can blank automatically on Android 15 and newer. That targets remote-help scams that ask you to share your screen while typing a password or CVV.
- Proactive call-forwarding protection. Android can block apps from quietly enabling call forwarding, and warn you before an MMI forwarding code typed on the dial pad takes effect. Google says this arrives in early December 2026 on Pixel and other select devices.
Verify unexpected requests through the bank’s official app or a number you already know, even if Android shows no warning.
Limitations worth knowing
- Enhanced protection is opt-in, not default.
- Cloud Gemini analysis is described for unknown senders. Do not assume a saved contact has been checked by this feature.
- Link Nano checks need newer flagship-class hardware than the base Enhanced detection list.
- Chat-notification alerts for WhatsApp and other apps are expanding via OEMs later this year; that is not the same switch as Messages Enhanced protection.
- No protection catches every slow-burn social engineering attempt. Your own pause before tapping a link still matters.
FAQ
Is Google Messages Enhanced Scam Detection available in India today?
Google announced a coming-weeks rollout and said some features may not be available in all countries. Check Google Messages settings on your phone after updating; do not assume every warning listed in the blog is live in India yet.
Does Enhanced protection send my chats to Google?
Google says the cloud analysis for Enhanced Scam Detection runs in Private AI Compute, with data encrypted and processed in a dedicated space it describes as inaccessible to anyone else, including Google. Separately, reporting a message as spam can send that message to Google for analysis.
Will my Xiaomi, vivo or OnePlus phone get scam alerts?
Google named those makers (and others) for expanding chat-notification Scam Detection later this year. Timing and exact models still depend on each manufacturer. Messages Enhanced protection itself is tied to AICore-capable devices and the Google Messages rollout.
Do I still need to be careful with UPI and KYC links?
Yes. These tools reduce risk; they do not remove it. Treat unexpected payment, KYC and courier links from unknown numbers as hostile until you verify them through an official app or known number.
