Cybercriminals take advantage of any communication channel in order to scam and steal our personal and banking data. Email, social networks, SMS, phone calls… Although more and more users are becoming aware of cyber scams, for some it can be difficult to differentiate a real call from a fraudulent one. Google wants to make it easy for us with its new “Scam Detection” function, which works with artificial intelligence.
Scam calls are very common in recent years. Although cybersecurity experts and the media try to raise awareness of this, and many users prefer not to pick up the phone when they call from unknown numbers, there are still those who fall into the trap.
If you answer the call and it turns out to be a scam, Google will be able to warn you in real time thanks to its AI. Recently, the company announced the launch of “Scam Detection”, a tool that is designed to protect Android users from telephone fraud attempts. Currently, the functionality is already rolling out in the public beta version of the English-speaking Google Phone in the United States. Therefore, the Google Pixel 6 or later will be the first to enjoy the novelty.
As this is an artificial intelligence function, it may fail at first and there may be false positives. Although the warning is useful, it will not be 100% reliable and, as it is used, it will be improved. In this way, users can rely on “Scam Detection” alerts and hang up as soon as possible without providing private data to cybercriminals.
How does Scam Detection work?
The ideal is not to answer fraudulent calls, but not everyone can ensure that a number is harmless before picking up. There are pages like ListaSpam that tell you when a number is suspected of being a scam. However, if we are waiting for an important call, we cannot waste time while the cell phone rings. In those cases, we can detect that the person on the other end of the phone is a cybercriminal by what they say, but it is complicated. That’s where Google’s artificial intelligence comes into play.
With “Scam Detection”, Google Pixel smartphones will be able to analyze conversation patterns and identify if it is a scam. It will do this in real time, so you’ll know if the AI is suspicious of the call while you’re chatting. For example, if the person on the other end of the line requests urgent transfers, or pretends to be an operator worker to ask for your personal information, alarms will go off.
As indicated by Google, the function capable of detecting telephone scams is only activated if the user so wishes, from the Phone app. By default, it is disabled, since there are Internet users who do not trust the use of artificial intelligence. However, the developers claim that the company does not spy on conversations and that all AI processing is done locally on the smartphone.
The AI model that Google has applied in “Scam Detection” to detect scams is Gemini Nano. Theoretically, this artificial intelligence is only present in the Pixel 9 series. But the technology firm has managed to implement the security function in previous models (from Pixel 6 onwards) through other robust machine learning models.
At the moment, the tool that identifies telephone scams in real time is only available in the United States and for conversations in English. Additionally, it has only reached Google Pixel devices. As it is perfected, the company may consider deploying it in more territories and in other languages with the aim of putting an end to call fraud.
