DIGITAL AGEISM PERSISTS IN ARTIFICIAL INTELLIGENCE-GENERATED IMAGES

Nicole Derenne, Lindsey Martens, Nicole Virgin, Phillip Hoffarth, Richard Van Eck, Marilyn Klug, Gunjan Manocha, Donald Jurivich · 2024

Read DIGITAL AGEISM PERSISTS IN ARTIFICIAL INTELLIGENCE-GENERATED IMAGES by Nicole Derenne, Lindsey Martens, Nicole Virgin, Phillip Hoffarth, Richard Van Eck, Marilyn Klug, Gunjan Manocha, Donald Jurivich (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11693346.txt.

Description

Nicole Derenne and co-authors check whether image-making AI still paints a biased picture of old age. Text-to-image systems often mirror social stereotypes, a pattern called digital ageism. The team fed the same set of prompts to the DALL-E 2 generator at two points a year apart. The prompts were drawn from terms used in aging care. They studied 164 images for perceived gender, race, class, and facial mood.

At both rounds, the images leaned heavily to white, middle-class men: white figures appeared five times as often as all other racial groups combined. The second round added more Asian figures, but gender and class gaps, and facial moods, barely moved. The authors conclude that bias in generated images of older adults persists despite public pushes for kinder views of aging. They note the study covers only one generator and one kind of content.

Catalog record

CREATORNicole Derenne, Lindsey Martens, Nicole Virgin, Phillip Hoffarth, Richard Van Eck, Marilyn Klug, Gunjan Manocha, Donald Jurivich
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE2 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11693346

Browse the full library · About the Tower of Babel