There are AI lie detectors being tested throughout EU airports right now. This is the biggest test of the system so far, and one of the biggest innovations in airport security. Installed in Hungary, Latvia, and Greece, the machines will ask tourists questions as they pass through customs.
With airport officials looking at different technologies to ease congestion, a variety of different methods have been tested throughout Europe, from face recognition technology, to lie detectors. Facial recognition features failed when the algorithm couldn’t decipher women and dark-skinned individuals. Critics of the lie detector feature are afraid that machines will inherently develop the same racist tactics as humans, as that’s already been shown in past demonstrations.
“The avatar will become ‘more skeptical’ and change its tone of voice if it believes a person has lied, before referring suspect passengers to a human guard and allowing those believed to be honest to pass through, said Keeley Crockett of Manchester Metropolitan University in England, who was involved in the project.”
The project, called iBorderCtrl, has so far only been tested on 32 people. Once it is installed and ready to go, scientists will use these trials to make sure the technology can work on its own. Until then, each AI lie detectors will be controlled by a human employee that can keep control of the situation, and make sure the new machinery is operating to plan. Keeley Crockett of Manchester Metropolitan University in England, who was involved in the project, explains, “I don’t believe that you can have a 100% accurate system.”
The AI robots “will collect data that will move beyond biometrics and on to biomarkers of deceit,” said project coordinator George Boultadakis, of information technology service company European Dynamics in Luxembourg.
“We’re employing existing and proven technologies – as well as novel ones – to empower border agents to increase the accuracy and efficiency of border checks,” adds Boultadakis.”
In addition to the ability to ask questions, the AI computers will require travelers to submit a photo of their passport, visa, and proof of funds. So far, for this test, only voluntary customers will undergo this questionnaire. However, if it works out, it’s likely that this technology will come to airports all over the world. If these trials go well, the UK, Spain, Poland, Germany, and Cyprus have all signed up to try them out next.
According to The Telegraph,
“The technology is advertised as having a ‘unique approach to deception detection’, analyzing the micro-expressions of travelers to figure out if the interviewee is lying.
Travelers deemed low risk during the pre-screening stage will go through a short re-evaluation of their information for entry, while higher-risk passengers will undergo a more detailed check.”
As technology contributes to security, these digital programs will likely end up being used to create a full-scale database. Police can then use that information to solve crimes.