A collection of affect recognition applications to reflect on black box systems

EmoCaptcha is a collection of speculative projects using affect recognition technology—human emotion detection—with escalating consequences to reflect on enigmatic black box systems and their capacity for misuse and discrimination. The absurd and unsettling integration of affect recognition in order to restrict access to video meeting spaces or adjust someone's social safety net payment in the COVID-19 pandemic world is meant to highlight the flawed science behind the affect recognition technology and the biases in the system's training data.

NYU ITP thesis archive site (with references)

emocaptcha.us (demo site)

Thesis Week presentation slides

EmoCaptcha [v1] Early EmoCaptcha prototype

EmoCaptcha Google Chrome Web Browser Extension [v2] EmoCaptcha Chrome Extension

EmoCaptcha Zoom Integration [v3] EmoCaptcha Zoom screenshots

EmoCaptcha COVID-19 Stimulus Payment [v4]

EmoCaptcha payment calibration EmoCaptcha payment amount EmoCaptcha payment verification



Advisors: Kathleen Wilson, Tom Igoe

Research sources: Clarinda Mac Low, Michelle Shevin, Larry Buchanan, Sara Watson, Caroline Sinders, Maya Indira Ganesh, Tom Igoe, Gilad Rosner

Headshot models: Andrew Lee, Cara Neel, Eva Philips, Dylan Dawkins, Tom Igoe, James Hosken, Khensu-Ra Love El

Special thanks: Deus Ex Dongle Thesis Class, EmoCaptcha user testers

© 2021 Adrian Bautista, built with GatsbyJS