Georgia Tech Researchers Aim to Increase Awareness of Emotion AI — By Letting People Try It
Posted July 29, 2025
Can you tell what someone is feeling based on their facial expression?
Proponents of emotion AI — a type of artificial intelligence that analyzes facial expressions, text, voice, and other cues to infer emotions — say it can do just that.
Georgia Tech researcher Noura Howell, who received an NSF CAREER award to study emotion AI in 2024, said the technology has a number of shortfalls that can lead to inaccurate results. Like generative AI, emotion AI is also subject to bias, and its use raises ethical and privacy concerns.
Despite these shortcomings, Howell said emotion AI has quietly shaped decisions in areas like hiring, education, mental health, and public safety in recent years.
Yet most people don’t know it exists.
Howell and Digital Media Ph.D. students Xingyu Li and Alexandra “Allie” Teixeira Riggs in the School of Literature, Media, and Communication are working to change that. They held workshops across Atlanta over the last two months, giving participants a rare opportunity to try emotion AI for themselves — and then share their impressions, ideas, and concerns.
Related Media
Xingyu Li (left) demonstrates the emotion AI system created for the team's workshops. The system has captured and analyzed her facial expression on the screen.
Related Links
Contact For More Information
Stephanie N. KadelIvan Allen College Communications