AI. From November virtual guards at the airport
Software that allows you to identify who is lying
At the airports in Hungary, Latvia and Greece a new potential security aids will be tested during November. In fact a virtual guard will interrogate incoming passengers from extra-EU flights and it will be able to understand if they are telling the truth or not. The Artificial Intelligence (AI) will go to support the agents who work to check documents confirming the name, age and date of birth, verifying the photo of the passport and asking questions to understand the purpose of the trip and how it was financed.
Practically an aid that could also make the figure of the man useless, since the virtual guard can also make more detailed investigations on the person in case he has suspicions. In this first test phase it will be used only on voluntary passengers -who will have to give their written consent during the flight-, in an important experimentation phase for "iBorderCtrl" costing so far 4.5 million euros. At the moment in the "laboratory" tests it has identified 85% of the lies of 32 people, but also the technicians involved in the project are skeptical about reaching 100%, and they could also create "racial" and "gender" problemes, like the IBM and Microsoft facial recognitions software that are more easily mistaken where they have to control women or people of color. The experimentation phase that will begin this month, however, will allow us to see whether or not the future of post-flight checks will be artificial intelligence.
© World Aeronautical Press Agency Srl