I’m the techie in my environment and that means my parents ask me technology questions virtually every time they see me, but friends and close family also take any opportunity to ask about possible purchases, problems with their devices and also with something that unfortunately happens more and more often: scam attempts and spam calls.
Those WhatsApp messages from a family member who needs a favor (special mention to this son’s scam), or the one with the likes, the one with the six-digit code…The list of potential scams is long and growing every day. And that’s without counting commercial calls: it doesn’t matter if the law has come into force, they use prior consent.
That’s why at this week’s Google I/O, when I saw the Gemini AI-powered scam call detector implemented in Android 15, I breathed a sigh of relief. Yes, Google’s call filter is wonderful, but with AI it’s even better. My question now is, Why doesn’t Gemini protect the phone completely?
Gemini call scam detection is good, but insufficient
Brief notes on how this technology works. Gemini Nano will be the on-device AI (it runs on the device, an important issue for privacy) of Android devices and applied to calls it will work as an evolution of the current call filter if we want (it will be optional). As soon as it is received, the artificial intelligence will listen to the content and at some point the call will stop and Android will issue a warning. Here we will have two options: either ignore the warning and continue the call, or hang up. It will work with both banking scams and other personal information such as PIN or password.
The function promises, but does not live up to it. Between the fact that we use technology more and more, the proliferation of hooks that invite us to fall for scams and that these are more elaborate, the reality is that This is neither the first nor the last time that they even make me doubt
Plus, I can’t always help them (urgency is usually an incentive), so beyond urging them to always say no when in doubt, this Gemini thing with calls seems wonderful to me. But This will only be the tip of the iceberg of all threats.. And of course, not everyone has someone on their side to deal with the problem and the time to take action. Tech-related scams are a big problem.
The idea of having a local artificial intelligence that monitors possible scams on calls is good, but it is insufficient: we find these fraud attempts in applications like WhatsApp, on the web, on social networks… So I say it loud and clear: Spam detection in Gmail and the call filter are not enough, but what Google is going to implement with Gemini is not enough either. And it won’t reach all devices either. In fact, it will probably first be something exclusive to Pixels and should become the norm. And in the meantime, what?
So Google, what Gemini Nano works on calls should be just the beginning, especially given the Mountain View company’s massive ecosystem of apps and services. No one has it easier: in the App Store, in your browser, in your office suite, in your email, etc. This is the type of use of artificial intelligence that can really make a difference and counteract the other side of the coin: AI that will be used to scam us more and better.
Cover | Montage with screenshot of Google I/O and photo by Iván Linares (Google Pixel 8 analysis)
In Xataka Android | I’ve already tested Android 15 beta 2: this is what I liked most about the new update
