Agonistic Image Generation: Unsettling the Hegemony of Intention

Current image generation tools largely follow an intention-centric paradigm that aims to actualize user intentions but neglects the sociopolitical conversations that these intentions are embedded in. As these tools become cornerstones of the media landscape, however, it is increasingly evident that sociopolitical conflicts over visual representation are inescapable parts of image generation. For instance, in March 2024, Google’s Gemini faced criticism for inappropriately injecting demographic diversity into user prompts. Although Gemini challenged user intentions, its top-down imposition of a standard for “diversity” ultimately proved counterproductive. In this paper, we present an alternative approach: an image generation interface designed to embrace open negotiation along the sociopolitical dimensions of image creation. Grounded in agonistic pluralism (from the Greek agon, meaning struggle), our interface actively engages users with competing visual interpretations of their prompts. Through a lab study with 29 participants, we evaluate our agonistic interface on its ability to facilitate reflection, or engagement with other perspectives that challenges dominant assumptions. We compare it to three existing paradigms: a baseline interface that emulates current image generation tools, a Gemini-style interface that produces “diverse” images, and an intention-centric interface that suggests aestheticized prompt refinements. We find that the agonistic interface enhances reflection across multiple measures, but also that reflection depends on authentically grounding diversity in relevant political contexts. Our results suggest that diversity and user intention need not be treated as inherently opposing values. Instead, interfaces can productively navigate tensions between competing perspectives, enabling users to engage with and evolve their intentions in meaningful ways.