Faception facial personality profiling
Released: May 2016
Faception is a Tel Aviv-based company that uses computer vision and machine learning to analyse facial images, identify personality types, and people in real-time.
Founded in 2014, Faception drew controversy in 2016 when it claimed to be able to identify personality traits and the ability to spot criminals, paedophiles, white-collar criminals and others, by analysing people's faces.
The system worked by collecting images of people's faces from photographs and video, and encoding and mapping their facial features on to a set of 15 proprietary 'classifiers', including academic, terrorist, and bingo-player.
Faception's claim that it could typically classify people with around 80 percent confidence were skeptically received by computer scientists, who reckoned the technology is not yet sufficiently good.
Others highlighted the ease with which the system could be used for covert surveillance and other purposes, and cited a contract that Faception itself had signed with an unnamed homeland security agency.
Sector: Business/professional services; Banking/financial services; Govt - police
Purpose: Identify personality type; Predict behaviour
Technology: Computer vision; Behavioural analysis; Emotion recognition; Facial recognition; Personality analysis; Machine learning
Issue: Accuracy/reliability; Bias/discrimination - race, ethnicity, gender; Ethics
Transparency: Governance; Black box