SÃO PAULO – Facial recognition technology comes loaded with comfort, yes, promises of greater security, okay. But it also comes loaded with a major, yet often ignored inconvenience: Violation of Privacy.

In parallel, the expansion of an entire security industry revolves around it and slowly but surely society turns into the Orwellian nightmare of a place of controlled citizens.

Derived from artificial intelligence, facial recognition took its first steps in the mid-sixties.

Those early attempts to use a computer to recognize a human face have led to a technology that has reached an amazing level of maturity.

Can we trust that companies will not market our faces?

Proof of this is the iPhone X, which enshrines something that a few years ago belonged to the domain of science fiction: unlock a phone with the image of our face.

“When you find a technology like this in a consumer device like the telephone,” says Enrique Dans, professor of Innovation at IE Business School, “it means that with it you can do everything”.

In China, a country that has set itself the goal of becoming a leader in artificial intelligence research and applications in 2030, one can now scan the face with the Xiaohua Qianbao mobile application and ask for a loan from the virtual bank operated by Xiaohua; go to a Kentucky Fried Chicken in the city of Hangzhou and pay with a smile -Smile to Pay is the latest system developed by the online payment application Alipay, and control student class attendance at Nanking University of Communications.

There, the technology of facial recognition advances firmly of the hand of Face ++, a Chinese start-up that at the end of October defeated  Facebook, Google, and Microsoft in tests of recognition of face images in the International Conference of Vision by Computer celebrated in Italy.

That same month the company raised 460 million dollars in a round of financing.

But the expansion of the phenomenon is not limited to that territory. Toronto stores use it to detect thieves. Facebook has been using it for a long time to tag the person who appears in the photo.

In fact, in 2015 it already announced that he could identify someone with an 83% success without showing his face; body type, hairstyle or posture are sufficient elements.

The new challenge for researchers is now to get the identification of people wearing dark glasses, veil, mask, balaclava: at the University of Basel, Switzerland, Professor Bernhard Egger works on a system that creates a 3D face pattern form of the uncovered areas of the face.

It all comes down to money, lots of it

Thus, the facial recognition market moves more than 3,300 billion in the world and could reach 7,7 billion in 2022, according to the consultancy MarketsandMarkets.

Banks, airlines, telephony, computer manufacturers, all open up to this new form of biometric identification that supposes a leap forward in front of the fingerprint and the iris.

Tace is not the same as the fingerprint. When we go to renew a bank card, we agree to give this biometric data to the authorities. But our face can be captured by whoever wants without our consent. With any camera on the street, anywhere.

This technology consists of two basic modalities, as explained in a telephone conversation from Michigan by the great expert Anil K. Jain, professor of computer engineering and director of the biometric research group at the University of Michigan.

The face detection or authentication, in which the system compares two images: the one we have stored in the phone -in the case of the iPhone- and a 3D model that is created from the face that is presented in front of the screen, and the face search, in which an image is crossed with those stored in a database to see that they match -to identify strangers-.

“In this second it is much easier to make mistakes,” explains Jain. “You need powerful computers and ample databases with millions of faces.”

It is this second modality that has triggered the debate on privacy and freedoms.

Its combination with the growing self-exposure in social networks is ending the era of anonymity.

The best example is offered by the FindFace application, which last year unleashed a storm in Russia: one pulls out the phone and takes a picture of the passenger traveling in the metro in front; the algorithm of the application crosses the image with those existing in the social network Vkontakte, which has more than 400 million profiles; with an efficiency of 70%, and lets people know who that person is.

The problem is where our face ends

The British newspaper The Guardian reported that the Australian Attorney General has been in talks with telephone operators and banks for the private use of its facial verification service in 2018.

Data protection experts are worried about the use that companies can make of their clients’ faces.

An investigation by The Washington Post revealed in November that Apple was sharing face information with some applications; and that made a change, as a result of journalistic research, requiring an app to inform users of the privacy policy.

Facebook, Google and Snapchat, on the other hand, are three of the companies that have already been sued in Illinois for the capture and storage of users’ images without their consent.

Can we trust that the companies of the new digital economy will not market our faces?

The Culture of Surveillance

“The problem is that there is a total lack of transparency,” says Kelly Gates, a professor at the University of California San Diego and author of Our Biometric Future: Facial Recognition Technology and the Culture of Surveillance.

“The police, like the Army, is experimenting, but we do not know what they are doing.”

This researcher, who now studies the techniques of forensic video analysis, assures that there is a proliferation of videos and data coming from drones, street cameras and stores whose analysis is outsourced to private companies:

“Scientists say that it is a technology with which many mistakes are made. There is no science that supports it and, even so, it continues to be used. ”

Everything is because it does not happen as in that dystopia signed by Terry Gilliam, Brazil, a 1985 film in which a mistake of data leads to the arrest of Mr. Buttle when the target was Mr. Tuttle.

“You are pursuing a perfect security that is never going to be achieved. To think that, in contexts of violence, all this is the great solution is like buying more air conditioners to solve the problems posed by climate change “.

In the end, the question is in which hands falls the use of this technology and the handling of our data.

Countries with problems of human rights and freedoms have a tremendous instrument for persecution of dissidents.

Does anyone imagine this technology in the hands of a far-right government in Europe? Or in a country ruled by radical Islamists?

We cite these cases because it is clear that in western countries governments are already using it to illegally spy on people everyday.

Leave a Reply

Your email address will not be published. Required fields are marked *