Scientists discover lie detector that uses artificial intelligence to detect micro-expressions

WION Web Team
Jerusalem Published: Nov 21, 2021, 09:08 PM(IST)

In this file photo, a representative image of Artificial Intelligence can be seen. Photograph:( Twitter )

Story highlights

Researchers believe it could make its way into the private sector, for example, to screen insurance claims or job applicants

Scientists have discovered a new lie detector that can read facial muscles that people won't even know they are using.

The study, conducted by the researchers at Tel Aviv University, has been in 'Brain and Behaviour.'

It was conducted on the basis of micro-expressions that vanish in 40 to 60 milliseconds due to which accuracy and speed played a key role.

Also read | Experts look to recycle dangerous space junk into rocket fuel in Earth's orbit

According to behavioral neuroscientist Dino Levy, ''it's not perfect, but much better than any existing [facial recognition] technology.''

''Since this was an initial study, the lie itself was very simple,'' he added.

Researchers tested the give-away indicators on 48 participants during which people had tells such twitching of eyebrows or cheek muscles.

''We successfully detected lies in all the participants and did so significantly better than untrained human detectors,'' explains Levy.

Also read | Facebook advertisers promoted false claims about climate change during COP26 summit

''Interestingly, individuals who were able to successfully deceive their human counterparts were also poorly detected by the machine-learning algorithm,'' he added.

The lie detector was driven by artificial intelligence and hopes to bring more transparency. It can also be used to ramp up border security.

Researchers believe it could make its way into the private sector, for example, to screen insurance claims or job applicants.

They usually face criticism from human rights groups that question the technology’s ability to accurately assess people’s intentions and its potential for discrimination.                                                                                   

(With inputs from agencies)

Read in App