4 Little Trees (4LT) student emotion recognition

Occurred: February 2021 

Can you improve this page?
Share your insights with us

An AI programme that analyses students' emotions as they learn in order to help teachers make distance learning more engaging and personalised, came under fire from researchers and digital rights activists fearful that emotion recognition technologies are intrusive, can be misused, and may be biased against people with darker skins.

Find Solution AI, the company behind 4 Little Trees, said its algorithm is 85% accurate and identifies happiness, sadness, anger, surprise, fear, and other emotions by analysing facial muscular micro-movements in real-time. Launched in 2017, 4 Little Treesand proved successful, with the number of schools reputedly using the system growing from 34 to 83 during the COVID-19 pandemic

However, a February 2021 CNN report cited researchers and activists noting that emotion recognition technologies are often intrusive, can be misused, and may be biased against people with darker skins. They also struggle to identify more complex emotions such as enthusiasm or anxiety.

A May 2021 Financial Times article noted that research into emotion recognition systems suggested that while they might be able to decode facial expressions much of the time, what a person is really feeling or thinking, or what they plan to do next, may be quite different.

Databnk

Operator: True Light College
Developer: Find Solution AI
Country: Hong Kong
Sector: Education
Purpose: Identify & monitor emotions
Technology: Emotion recognition; Facial analysis; Gesture analysis; Computer vision
Issue: Accuracy/reliability; Privacy; Surveillance; Bias/discrimination - race, ethnicity
Transparency: Governance; Black box