Robots made with the help of Artificial Intelligence (AI) technology for some time (Robots) are very much discussed. Their ability was praised to such an extent that it was even said that one day they would take the place of humans. But now these robots have flaws (Artificial Intelligence Robots) are being seen. These robots are discriminating between men and women as well as black and white. This information has come to the fore in research done by John Hopkins University, Georgia Institute of Technology and Washington University.
It has been found that the reason behind robots learning conservative things is poor neural network model. They are merely estimating their jobs according to the color of the people. According to robots, if someone’s complexion is black, then they are blue collar workers. They are giving more importance to men than women. They are being called powerful in the society. Regarding this research, author Andrew Hundr, a postdoctoral fellow at Georgia Tech, said that we are taking the risk of creating a generation of racist and sexist robots.
Internet data is used
According to researchers, scientists who build robots use Internet data to create AI programs. The data that is downloaded from the Internet also contains biased content, which is later incorporated into new algorithms as well. Due to which the robots start adopting a biased attitude. Researchers have done experiments to test this behavior. They were investigated with the help of human faces, in which the robots were asked who was the criminal, doctor and home maker. It was found that robots gave 8 percent more preference to men than to women.
Researcher Vicky Jango says that if this continues, then this discrimination will reach people’s homes and can affect them. If the child of the house asks the robot to bring a doll, then the robot may give only a white doll. With this, discrimination will start among children regarding color and race.
Get the latest world news updates