Google announced an opt-in Enhanced Protection mode for Google Messages on October 14. It adds cloud-based Gemini analysis on top of the app's existing on-device scam detection, aimed at scams that build trust gradually before asking for money or credentials.

Google says the cloud analysis runs through Private AI Compute for conversations with unknown senders. A separate link check uses Gemini Nano on compatible phones, including Pixel 10 and newer and recent Galaxy models, to assess unfamiliar links against real-time web signals.

The mode is not yet generally available. Google says it will roll out in the coming weeks, which is why this is listed as announced rather than released.