AI protections for Android users have arrived, but should you use them?
As consumer Android devices continue to come under attack from those who would defraud you, the timing from Google couldn’t be better: AI-powered scam protection for calls and messages on your Android smartphone has just rolled out. The question is, seeing as this feature is enabled by default for messages at least, what are the privacy protections and should you disable it?
Google Announces New AI-Powered Scam Detection For Android
Designed to protect users from the increasingly sophisticated and damaging phishing attacks that have been plaguing users over recent months, the new features target what are known as conversational scams. The types that are seemingly harmless to begin with, but quickly evolve into malicious and harmful.
A joint Mar. 4 announcement by Lyubov Farafonova, product manager for Phone by Google and Alberto Pastor Nieto, senior product manager for Google Messages, RCS Spam and Abuse, has confirmed that these new AI-powered scam detection features for Android smartphone users are now rolling out.
Traditional spam and scam protections focus more on protecting users before a conversation starts, so unsurprisingly are much less effective against the kind of attacks that exploit social engineering tricks. “To better protect users,” Google said, “we invested in new, intelligent AI models capable of detecting suspicious patterns and delivering real-time warnings over the course of a conversation, all while prioritizing user privacy.”
Ah yes, privacy, I’ll return to that momentarily as the fact that these protections are AI-powered and one is enabled by default will undoubtedly be of concern to many readers. But first, let’s explore how the new protections actually work.
How The New AI-Powered Scam Protections Work
When it comes to scam detection for Google Messages, this use Google AI to “proactively address conversational scams by providing real-time detection even after initial messages are received,” Google said. So, when on-device AI detects any suspicious patterns, be they within SMS, MMS, or RCS messages, you will get a warning about a potential scam taking place along with the option to either dismiss or report and block the sender.
As for the call protection, this is an additional rollout to that which was first introduced late last year for U.S.-based and English-speaking Phone by Google public beta users on Pixel phones. This also leverages AI models processed on-device, analyzing conversations in real-time.”If a caller tries to get you to provide payment via gift cards to complete a delivery,” Google said, “scam detection will alert you through audio and haptic notifications and display a warning on your phone that the call may be a scam.”
As for the rollout timetable, Google Message AI protection is launching for English speakers in the U.S., U.K. and Canada, with other countries to follow soon. Call protection will initially be available, using Gemini Nano, to all English-speaking Pixel 9+ users in the U.S.
How Private Are The AI Protections From Google?
The question on the lips of many will be abashed around the fact that these protections involve an AI analysis and one is enabled by default. How private can this be, and do I need to disable these features, in other words? The biggest clue comes in something that has already been noted in describing how the protections work, namely through on-device scanning.
Scam detection on Google Messages only applies to conversations with non-contacts, and your privacy is protected as all message processing remains 100% on-device. “Your conversations remain private to you,” Google said, “if you choose to report a conversation to help reduce widespread spam, only sender details and recent messages with that sender are shared with Google and carriers.”
Much the same applies when it comes to call protections. “We built this feature to protect your privacy by processing everything on-device,” Google said; “Call audio is processed ephemerally and no conversation audio or transcription is recorded, stored on the device, or sent to Google or third parties.” The difference being that call protection is off by default.
So, should you disable message protection and not enable call protection? The answer to both, as far as I’m concerned, is no. This is one use of AI that I can get fully behind and so should you. I have reached out to Google for a statement regarding privacy concerns.