Nudging Creativity in Typography with Font Pairing Prompts

Graphic designers face an overwhelming abundance of font choices in the digital marketplace, leading to difficulties in creating innovative typographic pairings and a tendency to rely on familiar options. Current industry practices and typography education lack effective methodologies to foster creative and diverse font use. This thesis addresses this knowledge gap by introducing Font Flow, a web application designed to enhance typography creativity through randomized font pairing prompts. The proposed system leverages structured constraints to challenge designers, prompting them to work with unexpected font combinations that may initially appear incompatible. By engaging in this problem-solving process, designers develop a deeper understanding of typography principles and expand their creative range. Unlike existing AI-driven tools that optimize for efficiency, Font Flow prioritizes creative exploration, fostering the divergent and convergent thinking necessary for professional designers to maintain relevance in an AI-augmented industry. A comparative study between human-generated and AI-generated designs revealed that while AI models produce visually competent results only slight below the average human made, when place head-to-head according to four criteria to human origin designs, the human produced work won vastly more times. User feedback highlighted the website’s potential to refine typographic decision-making through structured constraints together along with experiences of surprise from misattributing the origin of work as either AI or human made. By integrating generative constraints into design pedagogy, this thesis contributes to the discourse on AI-assisted creativity, offering a model that enhances designers' ability to craft compelling, nuanced typographic compositions in a rapidly evolving digital landscape​.