Artificial Intelligence-Powered Voice Fraud on Rise

McAfee conducted a study , in which she found out that artificial intelligence technologies are increasingly used to cloning the votes of people and deceiving their loved ones. Last month, we talked in detail about one of these cases, now we will consider the general statistics.

McAfee researchers interviewed a little more than 7 thousand people from different countries, about 25% of whom said they had already encountered voice fraud. Of these, 77% lost money as a result of cunning deception.

In addition, experts analyzed the availability and effectiveness of tools for cloning voice using AI. Over a dozen free programs were discovered on the Internet that did not require much experience or knowledge for use. And in one of the tested tools, there was enough audio shopping with a lasting only three seconds to copy a voice with an accuracy of 85%.

Voice is a unique identifying sign of each person. In telephone calls, trust between familiar interlocutors is instantly established if they hear and recognize each other’s voice. However, with an increase in the popularity of AI, such a fundamental thing as trust, it has become much easier to manipulate.

Fraudsters use AI to cloning the voice and then send fake voice messages or call the victim’s contacts, in real time pretending to be helped. At the same time, 70% of the respondents who took part in the McAfee study are not sure that they could distinguish a fake from a real voice, especially if a calling fraudster pretends not to the closest acquaintance of the victim.

45%of respondents said that they would answer a message from a friend or a relative in need of money, especially if they thought that this was their second half (40%), a parent (31%) or a child (20%). And the most likely reasons for the request for help called the respondents are an accident (48%), robbery (47%), loss of a phone or wallet (43%), problems abroad (41%).

One way or another, the price of the error can be high: more than a third of the respondents affected by voice fraud lost more than $ 1,000, and 7% – from $ 5,000 to 15,000.

The survey also showed that due to the spread of diphs and other kind of misinformation, people became more careful about what they see on the Internet. 32% of respondents said that now they trust social networks much less.

“Artificial intelligence opens up incredible opportunities, but any technology has the potential of abuse in the wrong hands. This is what we see today: the availability and ease of use of AI help cybercriminals to scaling their actions in more convincing ways,” said Steve Grobman, Steve Grobman. Technical Director McAfee.

/Reports, release notes, official announcements.