Amazon robot takes 28 US congressmen for criminals - ForumDaily
The article has been automatically translated into English by Google Translate from Russian and has not been edited.
Переклад цього матеріалу українською мовою з російської було автоматично здійснено сервісом Google Translate, без подальшого редагування тексту.
Bu məqalə Google Translate servisi vasitəsi ilə avtomatik olaraq rus dilindən azərbaycan dilinə tərcümə olunmuşdur. Bundan sonra mətn redaktə edilməmişdir.

Amazon Robot Takes US 28 Congressmen for Criminals

Identification of the offender can be a difficult task, so it is not surprising that law enforcement agencies are turning to technology for help. Face recognition systems from sci-fi films have become a completely tangible reality, but their accuracy is still not ideal.

Photo: ACLU

In a new series of studies that activists from the American Civil Liberties Union (ACLU) decided to use for Amazon's Rekognition facial recognition system, photographs of members of the US Congress were used mixed with photographs of criminals, says New York Post.

Software that is used by many law enforcement agencies in the United States seems to have some very serious limitations, especially when using the default settings.

The ACLU organized a study using 25 000 public police snapshots, and then gave the task of Rekognition to compare these images with photos of all 535 members of Congress. When the results were ready, snapshots of the faces of the 28 legislators were positively identified as coinciding with the faces of known criminals.

“Members of Congress who were falsely matched to the criminal database include Republicans and Democrats, men and women, legislators of all ages, from across the country,” Union officials said.

The software is available for widespread use, and the ACLU said that the entire test cost them only a 12,33 dollar. But, although its accuracy is far from astronomical, Amazon representatives stood up for their face recognition system, saying that the American Union of Civil Liberties used the default configuration, not working with settings, to reduce the frequency of false matches.

“A confidence level of 80% is appropriate for recognizing hot dogs, chairs, animals and other objects in photos for social media, but it is not practical to identify people with the same degree of confidence,” Amazon said. The company recommends that law enforcement change the settings by at least 95%.

Still, the study does make one wonder: what are the chances that government officials are proactively and correctly configuring system settings rather than simply using the default configuration? If Rekognition is used in its original settings and law enforcement relies too much on the electronics, it could result in many innocent people being questioned or detained for crimes in which they were not involved.

Read also on ForumDaily:

How a smart home device turned family life into a home horror movie

American left on Amazon reviews of the weapons with which they killed their victims

6 things you can get for free from Amazon 

Amazon employees complain about working without air conditioning, exhaustion and dehydration

Miscellanea In the U.S. crime Amazon software
Subscribe to ForumDaily on Google News

Do you want more important and interesting news about life in the USA and immigration to America? — support us donate! Also subscribe to our page Facebook. Select the “Priority in display” option and read us first. Also, don't forget to subscribe to our РєР ° РЅР ° Р »РІ Telegram  and Instagram- there is a lot of interesting things there. And join thousands of readers ForumDaily New York — there you will find a lot of interesting and positive information about life in the metropolis. 



 
1084 requests in 1,150 seconds.