accessibility design

Current retail AI frontends commonly operate under an “ideal user” assumption, overlooking the perceptual, motor, and cognitive differences of users with disabilities and diverse needs, thereby creating systemic accessibility gaps. This study employs qualitative analysis of representative applications—including virtual assistants, virtual try-on systems, and hyper-personalized recommendation engines—to uncover interaction design mechanisms that inadvertently exclude atypical users. It identifies the root cause not in technical limitations but in procurement and organizational processes that lack accessibility mandates. To address this, the work proposes a “front-end assurance” framework that aligns the multimodal capabilities and intelligence claims of AI systems with the genuine diversity of end users, filling a critical gap in existing AI governance regarding user experience and advancing the practical implementation of inclusive design in commercial contexts.

LLM-driven conversational persona generators exacerbate ethical risks—including bias, manipulation, and societal unpredictability—due to their dynamic responsiveness and data dependency, rendering traditional conversational user interface (CUI) governance frameworks inadequate. This paper introduces the first interdisciplinary ethical governance framework specifically for LLM personas, integrating human-computer interaction, AI ethics, and sociotechnical practice perspectives to establish a dynamic accountability mechanism spanning design, evaluation, and co-evolutionary iteration. Methodologically, it combines ethical impact assessment, participatory workshops, and multi-stakeholder consensus building. Key contributions are: (1) an actionable ethical design guideline for LLM personas; (2) a structured persona evaluation checklist; and (3) an open-source design toolkit. The work advances initial standardization consensus between academia and industry on responsible CUI development, providing methodological foundations for transparent, inclusive, and user-centered LLM persona systems.

The human-computer interaction (HCI) field lacks a systematic conceptual understanding and practical guidance for technological fairness. Method: Through a systematic literature review (SLR) of 202 HCI papers addressing fairness, this study develops the first integrative analytical framework for technological fairness, employing thematic coding, cross-framework comparison, and qualitative meta-synthesis. Results: It reveals (1) a structured motivation map for fairness research; (2) a taxonomy of fairness definitions and frameworks spanning procedural, distributive, recognition, and relational dimensions; and (3) a scalable, theory-informed framework for fairness-oriented HCI research—functioning simultaneously as a conceptual map and methodological guide. This work bridges a critical gap in translating fairness principles into actionable HCI practice.

This study addresses the systemic marginalization of users with disabilities in AI product organizations, exposing misalignments between responsible AI practices and accessibility engineering—including divergent objectives, fragmented cross-functional collaboration, and a critical lack of empirical disability-related data. Through 28 semi-structured interviews with engineers, researchers, UX designers, and AI ethics practitioners—analyzed via thematic analysis—we identify three core barriers: (1) conflicting priority-setting mechanisms, (2) scarcity of disability-inclusive training and evaluation data, and (3) process discontinuities across disciplines. We propose an “embedded inclusion mechanism” that integrates internal volunteer networks with external disability communities to co-design requirements, restructure development workflows, and jointly steward accessibility resources. Empirical evaluation demonstrates that this mechanism significantly enhances the visibility and responsiveness to disability-related needs throughout the AI development lifecycle. Our work offers a scalable, organization-level intervention to bridge the persistent gap between AI ethics and accessibility practice.

Designers lack a bidirectional mechanism to translate design requirements into large language model (LLM) behaviors—and vice versa—hindering human-centered LLM integration in UX practice. Method: We propose the “Designer-Centric Adaptation” paradigm, emphasizing dynamic, human-in-the-loop shaping and responsive refinement of LLM behavior. Based on this, we developed Canvil—a Figma plugin enabling real-time parameter tuning, feedback-driven closed loops, and collaborative iterative design. Contribution/Results: Through design probes, co-design workshops with six designer pairs, and qualitative thematic coding, we demonstrate that designers effectively optimize both LLM adaptation strategies and interface designs using Canvil. We synthesize a cross-role collaboration workflow and establish the first systematic methodology for human-centered LLM adaptation—bridging the longstanding gap between UX design and LLM engineering, and advancing practice-oriented, human-centered LLM application development.

This study examines how organizations with mixed abilities navigate the diverse—and sometimes conflicting—accessibility needs of their members. Through six months of participant observation and in-depth interviews with a cross-disability nonprofit organization that originated as a blind sports group, the research reconceptualizes “accessibility conflicts” not merely as technical barriers but as critical opportunities to expose power dynamics, foster accountability, and enact restorative practices. The findings illuminate the dynamic interplay among technological constraints, role diversity, communication norms, and organizational objectives, offering a theoretically innovative and practically valuable perspective for inclusive design.

This study addresses the underexplored role of UI/UX designers in organizational privacy practices, a dimension often overlooked in favor of developer-centric approaches. Through semi-structured interviews with twelve privacy-advocating UI/UX designers and subsequent thematic analysis, the research systematically investigates their privacy-related perceptions, influencing factors, cross-functional collaboration challenges, and coping strategies. It reveals, for the first time, how designers navigate tensions among business objectives, technical constraints, and team dynamics through value-driven and adaptive approaches. The study identifies how individual characteristics and organizational contexts shape privacy advocacy, elucidates mechanisms of friction in interdisciplinary collaboration, and proposes designer-centered pathways for organizational change alongside actionable tooling recommendations. These findings offer both theoretical insights and practical foundations for fostering privacy-supportive design ecosystems.

This study addresses the limitations of current generative AI systems, which rely heavily on chat-based interactions and impose high cognitive demands that exclude individuals with intellectual disabilities—particularly in prompt formulation, information processing, and credibility assessment. To bridge this gap, the authors propose a dual-layer interaction framework integrating structural scaffolds (e.g., reliability indicators, context management) and experiential scaffolds (e.g., pacing control, multimodal guidance). Developed through a collaborative design approach combining computer science and industrial design, this framework expands the design space for cognitively inclusive AI. The work identifies core mechanisms—including initial calibration, proactive prompting, and direct manipulation of response segments—to formulate an interaction paradigm tailored to cognitively diverse users, thereby laying the groundwork for expert refinement and empirical validation.

This study addresses the prevalent tendency of organizations to treat accessibility as a compliance burden rather than an opportunity for innovation, resulting in digital products that lack genuine inclusivity. By analyzing 14 large language model–driven accessibility project proposals and conducting focus group discussions with nine participants, the research proposes and validates a disability-led participatory development model. Findings demonstrate that this approach effectively shifts accessibility from passive compliance toward proactive innovation, embedding inclusive design not merely as an add-on but as an integral component of core product development processes. Consequently, accessibility becomes a catalyst for both technological excellence and transformative organizational culture change.

This study addresses the transparency challenges in large language model–driven conversational recommender systems, which, despite their fluency, often undermine users’ understanding, trust, and control over recommendations. The authors design and implement a laptop recommendation chatbot featuring constraint-based generation, on-demand ranking explanations, and product comparison capabilities. Through a moderated think-aloud usability study, they empirically find that “design transparency” does not necessarily enhance user comprehension and identify ranking explanations as the most severe usability issue. The study further codes and prioritizes identified problems by severity, revealing that while users appreciate the system’s reduction of cognitive load, they strongly desire more direct manipulation controls. These findings offer critical design implications for human-centered conversational recommender systems.