Conference Agenda

As intelligent cockpits transition toward experience-centered scenario design, designers must organize increasingly complex relationships among users, behaviors, and system responses. While Generative AI (GenAI) accelerates early-stage exploration, it often struggles to interpret designers’ intentions, resulting in semantic drift and inconsistencies. This study proposes NarratAIve, a structured narrative–driven framework for human–AI co-creation in intelligent cockpit design. The framework encodes users, actions, contexts, and feedback into computable narrative units that serve as clear semantic inputs for cross-modal AI generation. Through two workshops with 24 participants (including graduate students and industry experts), six cockpit concepts were produced, demonstrating how narrative acts as a controllable mechanism for guiding AI and enabling iterative alignment between narrative text and AI-generated spatial, interaction, and formal expressions. The findings show that structured narrative significantly improves design clarity, enhances human–AI negotiation, and expands the breadth and precision of scenario-based concept development.

This study examines how generative AI can transform original characters (OCs) into interactive agents that improve character believability in design contexts. We propose a human–AI co-creation workflow that integrates dialogue, image, and context modalities, stabilizing style and behavior through structured prompts and expert feedback. In a mixed-method evaluation with five expert creators and a general audience, the approach increased perceived realism and emotional authenticity, while psychological complexity changed little—indicating the need for longer-horizon design. Dialogue contributed most to credibility, with context reinforcing coherence and images showing limited gains. We distill design principles for cross-modal alignment, style stability, and progressive complexity, offering a practical path to scalable, collaborative character design and more engaging interactive experiences.

In the wave of artificial intelligence development, employing computer technology to advance the innovative design of Chinese-style furniture and optimize the intelligent development of cultural products is crucial for revitalizing the industry and strengthening cultural confidence. This study identifies issues of homogenization and cultural dilution in intelligent furniture design and proposes an intelligent design model based on convolutional neural networks (CNNs) and a latent diffusion model (LDM). CNNs are used to recognize cultural features in Chinese-style furniture, while the LDM generates design schemes, rapidly producing Chinese chairs that meet contemporary aesthetic preferences while retaining traditional cultural attributes. The study enhances the intelligent recognition and application of fine cultural details, avoids weakened aesthetic value caused by single-algorithm approaches, improves the efficiency of intelligent design for Chinese-style furniture, and offers new perspectives for the creative development of cultural products.

Current food tracking applications are largely designed to support behaviour change among users who track their food intake to meet predefined goals. As a result, individuals who are curious about food tracking but do not have specific goals remain underserved, limiting the potential of these tools for preventive care. To address this gap, we conducted a three-phase interview study using an existing food journaling application as a probe to reimagine food tracking experiences for inexperienced users. Thirty-one participants used the app for 14 days and discussed their views on healthy eating, food practices, and their tracking experiences. Our analysis identified four distinct roles that food tracking applications can play to support inexperienced users during early engagement; each is tied to different communication styles and feature needs. We offer design implications that align with inexperienced users’ values, motivations, and learning needs, advancing more inclusive and preventive approaches to food tracking applications.

Artificial intelligence is bringing comprehensive impacts to the approaches, methods, and efficiency of creation. As a result, designers are being driven to tackle more untried tasks, rather than engaging in work they have long been proficient in. This study organised a co-design experiment involving novice designers and Textual GenAI to investigate the influence of prior experience and AI literacy on collaborative design outcomes. We conducted a detailed coding of collaborative activities in the design process. This uncovered the activity characteristics, shifts, and collaboration patterns in the co-creation process. These findings deepen the understanding of the Human-AI collaborative design process and the impacts brought about by prior experience and AI literacy. They also provide valuable insights for formulating adjustment measures when different types of designers collaborate with AI.

Research on AI and Intangible Cultural Heritage (ICH) often reflects a static, object-centered perspective. This paper instead explores how generative AI can be embedded in the living practices of ICH. Through a speculative intervention in a Huayao embroidery community in Hunan, China, we reframe AI not as a tool for content generation, but as a social mediator within local craft networks. By performing as an “AI Cross-stitch Pattern Merchant” during the local festival, the study reveals how AI’s creative empowerment of young women sparked intergenerational tensions around legitimacy, labor, and authority. While youth embraced GenAI for playful self-expression, elder embroiderers judged its outputs by communal aesthetics and moral hierarchies, revealing the moral economy shaping creative legitimacy. These dynamics show how after-AI practices can unsettle the domestication of women’s creativity in heritage. We argue for a shift from generation to negotiation, toward socially embedded and relational AI practices in ICH.