From job titles to jawlines: Using context voids to study generative AI systems
This work addresses the issue of bias and assumptions in generative AI for researchers and practitioners, though it is incremental as it applies a novel methodology to an existing problem.
The paper tackles the problem of studying generative AI systems' behavior under radical uncertainty by introducing a speculative design methodology that bridges unrelated domains to create context voids, using a case study of generating headshots from CVs with ChatGPT and DALL-E, and finds that the system produces biased representations and stereotypes.
In this paper, we introduce a speculative design methodology for studying the behavior of generative AI systems, framing design as a mode of inquiry. We propose bridging seemingly unrelated domains to generate intentional context voids, using these tasks as probes to elicit AI model behavior. We demonstrate this through a case study: probing the ChatGPT system (GPT-4 and DALL-E) to generate headshots from professional Curricula Vitae (CVs). In contrast to traditional ways, our approach assesses system behavior under conditions of radical uncertainty -- when forced to invent entire swaths of missing context -- revealing subtle stereotypes and value-laden assumptions. We qualitatively analyze how the system interprets identity and competence markers from CVs, translating them into visual portraits despite the missing context (i.e. physical descriptors). We show that within this context void, the AI system generates biased representations, potentially relying on stereotypical associations or blatant hallucinations.