Android is adding several scam protections, including a Gemini-powered check of message conversations and unfamiliar links. The changes also cover screen sharing, call forwarding and calls claiming to come from banks, giving users new ways to pause before a scam becomes an account takeover.
Watch Desk analysis
What happened
Help Net Security reports that Google is rolling out an opt-in Enhanced protection mode in Google Messages in the coming weeks. It uses cloud-based Gemini models through Private AI Compute to assess conversations with unknown senders for scam intent. On supported devices, Gemini Nano will also use real-time web signals to assess unfamiliar links before users open them. Google says the feature adds redesigned warnings; it does not describe these checks as a guarantee that every scam will be caught.
Other protections have different timelines. Android is extending screen-sharing safeguards to Chrome and Android WebView this month on devices running Android 15 or later, blanking login fields and payment screens during live sharing. Call-forwarding protection is due in early December on Pixel and select other devices. Android will warn users before an MMI forwarding code takes effect, and block apps from enabling call forwarding in the background. Google is also working with participating banks on verified financial calls: if a caller claiming to represent a bank fails verification, Android will end the call and alert the user.
Why it matters
The changes address several points where a scam can turn into a loss: a persuasive message, a dangerous link, a shared screen or a spoofed bank call. Gemini is being used as one layer of message and link assessment, while the other measures rely on device-level warnings and restrictions. That mix is more practical than expecting people to spot every red flag unaided.
Availability is not uniform. Some features are opt-in, some depend on supported devices, and verified calls require participating banks. The protections should be judged by what reaches which phones, not by the length of the announcement.
Our read
This is a substantial set of defensive changes, with a specific AI role rather than AI pasted onto the headline for decoration. The most useful test will be whether the checks are available across the Android devices and banking services people actually use, and whether warnings arrive early enough to change what they do.
What to watch
- Which devices and regions receive the Gemini-powered checks, and when.
- Which banks participate in verified financial calls.
- Whether Android publishes more detail on coverage and how the scam checks perform.
Discussion spark: Should scam protection rely more on AI interpreting messages and links, or on hard device-level blocks that stop risky actions outright?
Sources and evidence
- Source update (16 October 2026, 08:53 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.