IT West Ltd

From the blog

News from IT West

Lie Detector App

A UK company has developed a lie-detecting app that could be used interviews or other situations where high levels of honesty in (initially) completing forms is required.

Video Combined With Analytics

The London based company, called ‘Human’, founded by Yi Xu, a former investment banker and investment news TV presenter, has a team that includes a data scientist, a micro-expression coder and a psychologist. The company has developed a system that uses video from a mobile device (or CCTV camera) combined with analytics software that can examine a person’s face and thereby determine the most likely emotions being felt at that instant.

The company says that it is able to humanise technology to decipher emotion and characteristics and predict human behaviour. The machine learning aspect of the system is also thought to deliver a better and more accurate understanding of a human’s feelings, emotions, characteristics and personality, with minimum human bias.

How?

The system is able to use a phone video (for example) to capture 172,000 tiny points of an individual’s face, and to use those to read subliminal facial expressions live, and to convert them into a range of deeper emotions and specific characteristic traits in real time.

Why?

The obvious application is a kind of commercial lie-detection system and as a way of getting more from a person’s responses than what is actually said or written by them. The idea is that a person’s reactions to various questions could be more useful than what their answers are, particularly where understanding strengths, weaknesses, and true motivations are concerned. In short, the ‘Human’ system could help companies / organisations with anything from hiring staff to fraud detection to customer satisfaction analysis, and the technology can profile potential customers based on their personality, as analyzed by A.I.

According to ‘Human’, an app of this kind could have real-world applications in:

  • Recruiting and employee retention – finding out about a candidate’s personality, screening candidates by emotional intelligence, and increasing diversity with minimum human bias.
  • Financial fraud detection – insurance claims and loan applications fraud based on subliminal behaviour.
    Customer satisfaction analysis – getting beyond any financial motivation to customer engagement and getting a better understanding of customer experiences.
  • Sales prediction – profiling customer characteristics and behaviour by personality, and predicting purchase behaviour.
  • Security detection – although facial recognition in crowds is already being used, the ‘Human’ system could identify a face in a crowd and detect concerned emotions.
  • Professional sports intelligence – detecting potential players’ characteristics and personality and predicting mental and emotional status before games.
  • Dating EQ – quantify dating partner’s emotional intelligence, and profiling characteristics and personality with empathy level.

Not Just ‘Human’

The ‘Human’ company is not the only company working on new kinds of combined technologies focused on learning more about people. For Example:

  • Utah-based company, Converus, has a product called EyeDetect, which monitors pupil dilation in the human eye to detect truths and lies. The system boasts 86% accuracy – better than a human expert.
  • Researchers at the U.S. National Center for Border Security and Immigration at the University of Arizona and the U.S.
  • The Department of Homeland Security are testing Automated Virtual Agent for Truth Assessments in Real-Time (AVATAR) which is a kiosk based system where a virtual agent asks security questions, then alerts human agents when the kiosk detects lying.
  • Back in July 2016, Toronto startup NuraLogix developed their Transdermal Optical Imaging app software which is able to read different blood flow patterns in the face to reveal different human emotions and thereby detect truth or lies.

AI -The Big Difference

The addition of AI into the technology mix is the element that could help these kinds of technologies to rapidly increase in capabilities and in real-world value e.g. lie detection connected to AI smart glasses or to a video-conferencing system, that can enable detection to take place without anyone but the user knowing about it.

Consent Issues

Capturingand using footage is however likely to present some potential issues based around consent e.g. with GDPR, as well as issues about how responsibly and legally they could be used and monitored in a commercial setting, not to mention issues around privacy and security (storage of profiling results and data used in the systems).

What Does This Mean For Your Business?

So much of the workings of business and the many relationships with all stakeholders is based around contracts (verbal and written), conversations and behaviour that have to rely upon a large element of trust and judgement, without having access to the full picture of true emotions, motivations, personalities, and likely outcomes. These new technologies, supercharged by AI could add value to many different areas of business that are based around decision-making and screening. The result of being able to use them in an affordable and convenient format e.g. apps and easy-to-operate systems, could deliver new insights that could translate into significant competitive advantages.