Overview

Image search for inspiration during ideation relies upon algorithmic systems like Google Image, Pinterest, and Instagram. Despite its critical importance at an early stage of the design process, algorithmic image search in ideation is often excluded from the study of AI and creative work, powered by non-specialist platforms like Pinterest and Instagram that were not purpose built as design tools.

Ideation is a professional practice weighted with cultural burdens that have to date not been adequately researched. Design processes today reproduce and extend societal harms structured through intersecting inequalities. Images searched via algorithmic platforms during ideation have aesthetics, values and points of view that can marginalize others. Algorithmic images are ranked through systems that Chandra notes favor “power, privilege, gender, and cool”. Even photographs of seemingly innocuous visuals like decorations and recipe inspirations on Pinterest are weighed with a performativity to adhere to rigid gender norms, idealizing hyperfeminine domesticity. Online, these images have persistence and replicability.

This research project asks what are the implications for creative work of the constant unfurling of hierarchization and modulation of distorted access embedded in AI platforms? Endogenous algorithmic systems that distort and polarize images are the systems creatives rely upon for original creative inspiration. As our attention is focused within a narrower sphere, what does this mean for the future of inclusive design practices in creative work?

To understand how AI is changing this aspect of creative work, I will be conducting workshops with design teams in their studios in London and Berlin. As the focus of this research is not the team dynamics but understanding the work practice, conducting research with teams that already have rapport allows me to more productively confront the research problem, facilitating richer findings.

Project website