Google Messages is adding a new layer of defense against text scams, and it's rolling out now after being announced earlier this month. The feature, called Enhanced Protection, sits above the app's existing Standard Protection, which already blocks spam and suspected harmful links on Android.
What's different this time is the AI doing the work. Enhanced Protection's Scam Detection feature uses cloud-based Gemini models to read the context of a conversation with an unknown sender and work out what a scammer is actually trying to do, rather than just matching known bad patterns. Google says that's an upgrade over the on-device detection already built into Messages, which struggles more with brand-new or unusually sophisticated scripts.
To handle the privacy problem that comes with sending a text conversation to the cloud for analysis, Google is routing it through Android's Private AI Compute. That's a secure processing environment Google describes as private to the user and not accessible to anyone else, including Google itself.
The second piece is link protection. Here the work happens on the phone, using the on-device Gemini Nano model alongside real-time web signals to judge whether a link from an unfamiliar number is safe before you tap it. For now, that on-device link check is limited to the Pixel 10 series, Samsung's Galaxy S26 lineup, the Z Flip 8 and Z Fold 8, plus what Google calls "other select devices." Phones outside that list presumably get Scam Detection but not the local link analysis.
The third change is cosmetic but useful: Google has redesigned the warning that pops up when a number outside your contacts sends you a link, giving you more context and, more time before you can tap through. It's a small friction point aimed at the exact moment people get scammed, the split second between seeing a link and tapping it without thinking.
Enhanced Protection isn't a single on-off switch. Inside Google Messages, under Settings > Protection & safety, users now see three distinct options rather than a toggle, letting people choose how aggressive the scanning gets.
The rollout is showing up first in the Google Messages beta, version 20261020_00_RC00, so it hasn't reached everyone running the stable release yet. Given Google's usual cadence for Messages features, a wider rollout to non-beta users would be the next step, though the company hasn't given a timeline for that.
The timing isn't an accident. Text-based scams routed through SMS and RCS have become one of the more common ways people get phished on phones, and Google has spent much of 2026 folding Gemini into more corners of Android, from Android Earthquake Alerts to on-device malware scanning. Enhanced Protection is the latest example of Google trying to make that AI work invisibly, in the background, before a user ever has to notice something's wrong.
SDFiled by The Software Desk
Apps, platforms, operating system releases, security fixes and the features arriving in them.
More from this desk →
Be the first to comment
Join the argument. No password, just your email or a passkey.