AI辅助情绪板生成迭代交互系统设计：情绪板用法及痛点、生成式AI创意应用及局限、设计师掌控感与体验、创意支持工具评估原则

Professional designers create mood boards to explore, visualize, and communicate hard-to-express ideas. We present ImageCascade, an intelligent, collaborative ideation tool that combines individual and shared work spaces, as well as collaboration with multiple forms of intelligent agents. In the collection phase, ImageCascade offers fluid transitions between serendipitous discovery of curated images via ImageCascade, combined text- and image-based Semantic search, and intelligent AI suggestions for finding new images. For later composition and reflection, ImageCascade provides semantic labels, generated color palettes, and multiple tag clouds to help communicate the intent of the mood board. A study of nine professional designers revealed nuances in designers' preferences for designer-led, system-led, and mixed-initiative approaches that evolve throughout the design process. We discuss the challenges in creating effective human-computer partnerships for creative activities, and suggest directions for future research.

This study examines how integrating Artificial Intelligence (AI) into mood board development enhances the ideation process in graphic design education. The research aimed to understand how AI-supported visual exploration influences students’ originality, complexity, and goal alignment during creative concept development. A mixed-method approach was employed, combining quantitative rubric-based evaluation of student mood boards with qualitative thematic analysis of interviews and classroom observations to capture both performance outcomes and design thinking processes. The findings show that AI tools expanded visual exploration and improved conceptual clarity, yet their effectiveness depended on how critically students engaged with prompt iteration and keyword synthesis. Students who refined descriptive keywords and combined AI outputs with digital imaging achieved higher originality and coherence. Complexity increased when students generated multiple AI iterations from different visual angles, whereas goal alignment benefited from structured mind mapping informed by tone and manner. Qualitative results revealed six interrelated themes: idea exploration, AI assistance, visual curation, AI limitations, ethics and reflection, and implementation recommendations, highlighting the interplay between human judgment and computational generation. Overall, the study affirms that mood board development remains essential in guiding conceptual direction and visual storytelling in design education. AI serves as a creative collaborator that supports deeper ideation and expands the boundaries of visual experimentation in contemporary design learning.

… Fourth, to ground these five roles in practice, I present a second example of a mood board in interaction design. Finally, I discuss what HCI and interaction design can learn from how …

We present findings from a deployment of SketchStorm, a tool for designers that supports sketch in a central canvas, whilst streaming images relating to a search query around the periphery. Our overarching goal was to explore the potential for combining sketching and use of examples, two activities that are associated with design ideation. Initial interviews with designers suggested that a tool that supports encounters with non-designerly content, that supports awareness of what has already been collected, and that allows this content to be laid out, manipulated, and integrated into the process of working out of ideas, would be of value. A month-long deployment allowed us to examine these ideas in more depth, through 'research through prototypes in practice' (Keller et al., 2009). Our findings highlight two ways in which web-based images can be utilised. On the one hand, they can serve as examples and, where this is the case, encounters with them should be rich and memorable, and tools should support a range of actions such as triaging, annotation, and manipulation. On the other hand, images can be used to create a backdrop to on-going activity, so as to underpin serendipitous encounters. Where this is the case, enabling designers to engineer these encounters, so that they are framed by moments of idleness and latent goals, is key.

Design ideation is a prime creative activity in design. However, it is challenging to support computationally due to its quickly evolving and exploratory nature. The paper presents cooperative contextual bandits (CCB) as a machine-learning method for interactive ideation support. A CCB can learn to propose domain-relevant contributions and adapt their exploration/exploitation strategy. We developed a CCB for an interactive design ideation tool that 1) suggests inspirational and situationally relevant materials ("may AI?"); 2) explores and exploits inspirational materials with the designer; and 3) explains its suggestions to aid reflection. The application case of digital mood board design is presented, wherein visual inspirational materials are collected and curated in collages. In a controlled study, 14 of 16 professional designers preferred the CCB-augmented tool. The CCB approach holds promise for ideation activities wherein adaptive and steerable support is welcome but designers must retain full outcome control.

Visual metaphors, which are widely used in graphic design, can deliver messages in creative ways by fusing different objects. The keys to creating visual metaphors are diverse exploration and creative combinations, which is challenging with conventional methods like image searching. To streamline this ideation process, we propose to use a mind-map-like structure to recommend and assist users to explore materials. We present MetaMap, a supporting tool which inspires visual metaphor ideation through multi-dimensional example-based exploration. To facilitate the divergence and convergence of the ideation process, MetaMap provides 1) sample images based on keyword association and color filtering; 2) example-based exploration in semantics, color, and shape dimensions; and 3) thinking path tracking and idea recording. We conduct a within-subject study with 24 design enthusiasts by taking a Pinterest-like interface as the baseline. Our evaluation results suggest that MetaMap provides an engaging ideation process and helps participants create diverse and creative ideas.

Generative machine learning models provide opportunities to support design work in various parts of the design process. This study investigates how generative machine learning and large language models may play a part in creative design processes of ideation, early prototyping and sketching. A workshop was conducted in which design practitioners and design researchers developed design concepts for a provided design case, with the help of GPT-3. The findings point to three main themes, including i) the practical usefulness and limitations of the system in design ideation processes, ii) how the form of user interaction shapes users’ expectations of the system’s capabilities and potentials, and iii), how the broader discourse around AI both limits and enables how co-creative processes involving human and AI unfolds. The discussion outlines design implications and alternative framings of this kind of co-creative design practices based on post-human perspectives on design and technology use.

Graphic designers often get inspiration through the recombination of references. Our formative study (N=6) reveals that graphic designers focus on conceptual keywords during this process, and want support for discovering the keywords, expanding them, and exploring diverse recombination options of them, while still having room for designers’ creativity. We propose CreativeConnect, a system with generative AI pipelines that helps users discover useful elements from the reference image using keywords, recommends relevant keywords, generates diverse recombination options with user-selected keywords, and shows recombinations as sketches with text descriptions. Our user study (N=16) showed that CreativeConnect helped users discover keywords from the reference and generate multiple ideas based on them, ultimately helping users produce more design ideas with higher self-reported creativity, compared to the baseline system without generative pipelines. While CreativeConnect was shown effective in ideation, we discussed how CreativeConnect can be extended to support other types of tasks in creativity support.

Designers create inspirational mood boards to express their design ideas visually, through collages of images and text. They find appropriate images and reflect on them as they explore emergent design concepts. After presenting the results of a participatory design workshop and a survey of professional designers, we introduce SemanticCollage, a digital mood board tool that attaches semantic labels to images by applying a state-of-the-art semantic labeling algorithm. SemanticCollage helps designers to 1) translate vague, visual ideas into search terms; 2) make better sense of and communicate their designs; while 3) not disrupting their creative flow. A structured observation with 12 professional designers demonstrated how semantic labels help designers successfully guide image search and find relevant words that articulate their abstract, visual ideas. We conclude by discussing how SemanticCollage inspires new uses of semantic labels for supporting creative practice.

ABSTRACT This study presents the potential of live topic visualization in supporting creative dialogs during remote idea generation. We developed a novel Creativity Support Tool (CST) to explore the effects of the live topic visualization. The tool emphasizes the interdisciplinary knowledge background of participants. Using Natural Language Processing (NLP) and topic modeling, the tool provides users with a live visual mapping of the domains and topics being orally discussed. To understand the tool’s user perceived effects, we conducted evaluation sessions and interviews with participants (N = 10) from two different disciplinary backgrounds: design and bioscience. The findings show that live visualization of domains and topics supported self-reflection during individual and collaborative creativity and encouraged a balanced discussion, which can mitigate discipline-based fixation in ideation.

Abstract Generative Artificial Intelligence (Generative AI) is a collection of AI technologies that can generate new information such as texts and images. With its strong capabilities, Generative AI has been actively studied in creative design processes. However, limited studies have explored the roles of humans and Generative AI in conceptual design processes, which leaves a gap for human–AI collaboration investigation. To address this gap, this study attempts to uncover the contributions of different Generative AI technologies in assisting humans in the conceptual design process. Novice designers were recruited to complete two design tasks in the condition of with or without the assistance of Generative AI. The results revealed that Generative AI primarily assists humans in the problem definition and idea generation stages, while the idea selection and evaluation stage remains predominantly human-led. Additionally, with the assistance of Generative AI, the idea selection and evaluation stages were further enhanced. Based on the findings, we discussed the role of Generative AI in human–AI collaboration and the implications for enhancing future conceptual design support with Generative AI’s assistance.

… students in their design work. We conducted an exploratory study in the Design domain involving … that highlighted a measurable increase in creativity, supporting the efficacy of both the …

Generative Artificial Intelligence (GenAI) applications in artistic and creative domains have gained substantial attention of late. These intelligent interactive systems, shaped by innovations in Large Language Models (LLMs) and Vision Language Models (VLMs), are materially impacting digital creative domains. While initial work to understand this space has highlighted new models and architectures, we lack a holistic view of how interactive GenAI systems are designed for user interactions across various artistic and creative domains. In this paper, we present a systematic review of interactive GenAI system designs for art and creativity in the HCI literature (N = 189), and a detailed taxonomy of interaction paradigms with design components. We shed light on the communities of design focus and decompose the system interaction designs, mapping these characteristics to creative domains, user interaction patterns, GenAI technologies, detailing under-represented spaces, and future directions of designing interactions for GenAI creativity.

… The findings suggest that while AI can inspire creative and novel ideas, there are … designs. Finally, this research highlights the potential for Generative AI to support humans in creative …

… creativity to organize how recent advancements in GAI can support each creative category. We propose a conceptual framework of GAI towards transformational creativity, and identify …

Abstract Generative artificial intelligence (GAI) has emerged as an indispensable tool across various design disciplines, increasingly integrated into different stages of the design process. However, research on the effectiveness of human-AI collaboration in design remains limited, and the role of AI in the creative process is still underexplored. This study presents a single-factor between-subjects experiment involving 79 participants to evaluate the impact of varying degrees of automation in generative design tools on user experience and design efficiency. The independent variable is the degree of automation of the GAI tool, categorized into three levels: low, moderate, and high. The dependent variables include time, creative experience, and creative quality. Creative experience encompasses task load, perception, and creativity support, while creative quality pertains to aesthetic and usability outcomes. The results indicate that increasing the level of automation significantly reduces design time. Moderate levels of automation are most effective in lowering task load and balancing human-machine collaboration, whereas extreme levels of automation may be counterproductive. Although highly automated GAI can rapidly enhance visual aesthetics, its impact on stimulating designers’ creativity and critical thinking is limited. This research contributes to evaluating the effectiveness of human interaction with AI in the design process and offers insights for optimizing the application of AI in design.

… designers, this mutually supporting combination of CAD modeling and generative AI can … Foreseeably, these resources can contribute to creativity promotion along the different stages of …

Generative AI (GenAI) promises to revolutionize creative work, but its value is not universal. Using controlled lab settings with students and real-world tests with professional designers, our research shows that GenAI is a double-edged tool. In the initial brainstorming (ideation) stage, GenAI reliably boosts creativity for all users. However, in the execution (implementation) stage, whereas novice designers continue to benefit from GenAI’s assistance, expert designers encounter inefficiencies—spending significantly more time without improving creativity, because GenAI’s methods conflict with experts’ well-established routines. For firms, this means adoption strategies must be nuanced. GenAI delivers the greatest value when applied to brainstorming, early concept development, and work by less-experienced employees. In contrast, deploying GenAI in later-stage production tasks, especially with seasoned professionals, may reduce efficiency. Managers and tool designers should avoid blanket promotion of GenAI across all tasks and instead develop targeted adoption strategies that align with employees’ expertise and the stage of the creative process. By tailoring GenAI use, organizations can harness its creative potential while minimizing risks of counterproductive outcomes.

Abstract Generative AI algorithms that are able to generate creative output are progressing at tremendous speed. This paper presents a research agenda for Generative AI-based support for designers. We present examples of existing applications and thus illustrate the possible application space of Generative AI reflecting the current state of this technology. Furthermore, we provide a theoretical foundation for AI-supported design, based on a typology of design knowledge and the concept of evolutionary creativity. Both concepts are discussed in relation to the changing roles of AI and the human designer. The outlined research agenda presents 10 research opportunities for possible AI-support to augment the designer of the future. The results presented in this paper provide researchers with an introduction to and overview of Generative AI, as well as the theoretical understanding of potential implications for the future of the design discipline.

Designers often struggle to sufficiently explore large design spaces, which can lead to design fixation and suboptimal outcomes. Here we introduce DesignAID, a generative AI tool that supports broader design space exploration by first using large language models to produce a range of diverse ideas expressed in words, and then using image generation software to create images from these words. This innovative combination of AI-based capabilities allows human-computer pairs to rapidly create a diverse set of visual concepts without time-consuming drawing. In a study with 87 crowd-sourced designers, we found that designers rated the automatic generation of images from words as significantly more inspirational, enjoyable, and useful than a conventional baseline condition of image search using Pinterest. Surprisingly, however, we found that automatically generating highly diverse ideas had less value. For image generation, the high diversity condition was somewhat better in inspiration but no better in the other dimensions, and for image search it was significantly worse in all dimensions.

Studies of Generative AI (GenAI)-assisted creative workflows have focused on individuals overcoming challenges of prompting to produce what they envisioned. When designers work in teams, how do collaboration and prompting influence each other, and how do users perceive generative AI and their collaborators during the co-prompting process? We engaged students with design or performance backgrounds, and little exposure to GenAI, to work in pairs with GenAI to create stage designs based on a creative theme. We found two patterns of collaborative prompting focused on generating story descriptions first, or visual imagery first. GenAI tools helped participants build consensus in the task, and allowed for discussion of the prompting strategies. Participants perceived GenAI as efficient tools rather than true collaborators, suggesting that human partners reduced the reliance on their use. This work highlights the importance of human-human collaboration when working with GenAI tools, suggesting systems that take advantage of shared human expertise in the prompting process.

This paper investigates the potential impact of deep generative models on the work of creative professionals. We argue that current generative modeling tools lack critical features that would make them useful creativity support tools, and introduce our own tool, generative.fashion1, which was designed with theoretical principles of design space exploration in mind. Through qualitative studies with fashion design apprentices, we demonstrate how generative.fashion supported both divergent and convergent thinking, and compare it with a state-of-the-art text-based interface using Stable Diffusion. In general, the apprentices preferred generative.fashion, citing the features explicitly designed to support ideation. In two follow-up studies, we provide quantitative results that support and expand on these insights. We conclude that text-only prompts in existing models restrict creative exploration, especially for novices. Our work demonstrates that interfaces which are theoretically aligned with principles of design space exploration are essential for unlocking the full creative potential of generative AI.

Current generative AI systems primarily utilize a prompt–response interaction model that restricts user intervention during the creative process. This lack of granular control creates a significant disconnect between user intent and machine output, which we define as the “Agency Gap”. This paper introduces the Agency-First Framework (AFF), which combines cognitive engineering and co-active design approaches to formally define human-AI collaboration. This is operationalized through the development of ten Generative AI Agency (GAIA) Heuristics, a systematic method for evaluating agency-centric interactions within stochastic generative settings. By translating the theoretical layers of the AFF into measurable criteria, the GAIA heuristics provide the necessary instrument for the empirical auditing of existing systems and the guidance of agency-centric redesigns. Unlike existing assistive AI guidelines that focus on output-level usability, the AFF establishes agency as a first-class design construct, enabling mid-process intervention and the steering of the model’s latent reasoning trajectory. Validation of the AFF was conducted through a two-tiered empirical evaluation: (1) an expert heuristic audit of state-of-the-art platforms, such as ChatGPT-o1 and Midjourney v6, which achieved high inter-rater reliability, and (2) a controlled redesign study. The latter demonstrated that agency-centric interfaces significantly enhance the Sense of Agency and Intent Alignment Accuracy compared to baseline prompt-response models, even when introducing a deliberate increase in task completion time—a phenomenon we describe as “productive friction” or an intentional interaction slowdown designed to prioritize cognitive engagement and user control over raw speed. Overall, the findings suggest that the restoration of meaningful user agency requires a shift from “seamless” system efficiency towards “productive friction”, where controllability and transparency within the generative process are prioritized. The major contribution of this work is the provision of a scalable, empirically validated framework and set of heuristics that equip designers to move beyond prompt-centric interaction, establishing a methodological foundation for agency-preserving generative AI systems.

… writings on human-AI interaction in co-creation settings, we aim to explore the responsive behavior of all actors and potential beneficiaries involved in both the co-creation process and …

Control is a critical yet underexplored concept in human-AI co-creativity and more broadly human-AI collaboration, where AI systems are expected to act as collaborative partners with creative autonomy. Existing frameworks for characterizing control remain limited and often fall short in capturing the tensions and complexities of co-creation dynamics. In this paper, we examine how experts conceptualize control and expect human-AI control dynamics by leveraging a recent framework on characterizing control as our theoretical probe. We conduct a semi-structured focus group study with nine experts in HCI, co-creativity, and AI. Our findings reveal that control is widely viewed as a dynamic, context-dependent construct that should adapt across different phases of co-creation, domains, and levels of trust in AI. Drawing on our findings, we propose a conceptualization of control along with actionable design implications for designing such AI systems. This work contributes to the literature on Human-AI collaboration, Computational Creativity, and HCI, advancing our understanding of control in co-creative human-AI partnerships.

Generative AI has greatly transformed creative work in various domains, such as screenwriting. To understand this transformation, prior research often focused on capturing a snapshot of human-AI co-creation practice at a specific moment, with less attention to how humans mobilize, regulate, and reflect to form the practice gradually. Motivated by Bandura’s theory of human agency, we conducted a two-week study with 19 professional screenwriters to investigate how they embraced AI in their creation process. Our findings revealed that screenwriters not only mindfully planned, foresaw, and responded to AI usage, but, more importantly, through reflections on practice, they developed themselves and human-AI co-creation paradigms, such as cognition, strategies, and workflows. They also expressed various expectations for how future AI should better support their agency. Based on our findings, we conclude this paper with extensive discussion and actionable suggestions to screenwriters, tool developers, and researchers for sustainable human-AI co-creation.

While automation replaces human labour in structured and repetitive tasks, augmentation enhances human capabilities through artificial intelligence (AI) collaboration. This article develops a typology of human–AI value cocreation, distinguishing between automation and augmentation capabilities across varying levels of task complexity. The proposed framework is structured along two dimensions: mode of technology engagement (automation vs augmentation) and complexity of task-in-context. The four resulting models of value creation are analyzed in terms of their temporal, spatial and hierarchical complexity, which shapes human–AI interaction in performing a given task. Temporal complexity relates to sequencing, timing and feedback loops; spatial complexity refers to physical environments and the integration of dispersed resources; and hierarchical complexity captures interdependence and interaction across micro- and macro-level structures. By introducing this structured analytical lens, the article contributes to a more holistic understanding of AI-enabled value creation. The findings may also inform a research agenda for future inquiry.

While Generative AI (AIGC) reshapes creative production, there is a growing need to understand user experiences beyond functional utility. Drawing on meaning-making theory and self-determination theory, this study explores the relationships between human–AI interaction factors and perceived meaning—the extent to which users find personal relevance and expressive alignment in generated content. Survey data were collected from 299 AIGC users and analyzed using a mixed-method approach combining partial least squares structural equation modeling (PLS-SEM) and fuzzy-set qualitative comparative analysis (fsQCA). PLS-SEM results indicate that system interactivity is positively associated with perceived meaning not only directly but also indirectly through a sequential pathway involving perceived control and psychological ownership. Similarly, post-editing behavior is linked to meaning primarily through these psychological factors. Furthermore, fsQCA identifies diverse configurational paths; while task openness shows no significant linear relationship, it emerges as a critical condition in specific combinations associated with high perceived meaning. Theoretically, these findings highlight the role of the control-ownership pattern in virtual co-creation. Practically, they suggest that AIGC design should prioritize controllable co-creation, supporting user agency to foster stronger content ownership and meaningful experiences.

The creativity exhibited by generative artificial intelligence (AI) has caused anxiety among some designers. This generative ability has a tremendous influence on the creative activities of designer. Therefore, this study aims to explore the process of co-creation between designers and generative AI, investigate the impact of generative AI on designers' creative activities and its underlying mechanisms, and explore how designers utilize agency to respond to this new challenge. Based on the theory of creative segment, a human-AI co-creative segment model is proposed to elucidate the mechanism of AI-augmented design. An observational study was conducted on a workshop where designers and generative AI collaborated in creating a design. Through analysing designers' cognitive behaviour during this process, their agency was identified, and three interactive modes of human-AI co-creation were proposed. Based on the above analysis, this study reflects on the current state of designer design and AI-augmented design tools, proposing that AI and designers should evolve collaboratively, and designers should exert their agency when facing new technologies like AI. Relevant tools and research should also aim to facilitate this process.

With the rapid advancement of generative AI technologies, the collaboration between designers and generative AI during the conceptual design stage has fostered a novel paradigm of Human-AI Co-creation in design. However, relevant systematic reviews remain relatively scarce. This study analyses 80 research papers and publications from prominent databases such as Web of Science (WOS), Scopus, ScienceDirect, and Google Scholar, based on rigorous selection criteria and utilizing the CiteSpace bibliometric analysis tool. We developed a comprehensive systemic model for Human-AI Co-creation Design: cognition, process, methodology, and outcome. Key findings encompass the transformation of design cognition from designer-centric to Human-AI collaborative cognition, the shift of design process from experience-driven to constraint-driven, the evolution of design method from unimodal to multimodal interaction, and the indirect influence of generative AI on design outcome. Furthermore, this paper discusses the boundaries of Human-AI Co-creation Design with generative AI and envisions future research directions for multidisciplinary collaboration and the future models of AI and designer creative fusion. Through a systematic analysis and summary of the current paradigm shift, this paper identifies the limitations of existing research and provides insights into future research directions for building a more efficient Human-AI Co-creation design paradigm.

… and agency of humans in AI-enabled decision-making processes. Section 3 introduces the co-creation methodology and the human-AI … specific co-creation design patterns for human-AI …

Generative AI agents have been widely used in creative activities. Complex creations in the real world require multiple interactions with agents to approach the creative goal gradually. However, there are currently no well-defined frameworks to guide human-agent co-creation in complex creative tasks. In this study, we propose GenFlow, a human-agent co-creation framework, to enhance the generation quality and user experience by providing explainable and controllable mixed-initiative with different granularities at different phases of creation. We demonstrated a GenFlow-based picture book agent and conducted an early user test with 11 participants. The results proved the framework’s effectiveness and usefulness in guiding users’ collaborative and interactive behaviors. Our research emphasizes the importance of explainable and controllable interfaces in human-agent co-creation and proposes design strategies. These strategies contribute to designing AI agents that can effectively support the creation process.

… Using a co-creation design workshop that involved 18 participants, the research … and a role-boundary framework for human-AI co-creation in the conceptual design phase, providing …

It has been increasingly recognized that effective human-AI co-creation requires more than prompts and results, but an environment with empowering structures that facilitate exploration, planning, iteration, as well as control and inspection of AI generation. Yet, a concrete design approach to such an environment has not been established. Our literature analysis highlights that compositional structures—which organize and visualize individual elements into meaningful wholes—are highly effective in granting creators control over the essential aspects of their content. However, efficiently aggregating and connecting these structures to support the full creation process remains challenging. We, therefore, propose a design approach of leveraging compositional structures as the substrates and infusing AI within and across these structures to enable a controlled and fluid creation process. We evaluate this approach through a case study of developing a video co-creation environment using this approach. User evaluation shows that such an environment allowed users to stay oriented in their creation activity, remain aware and in control of AI’s generation, and enable flexible human-AI collaborative workflows.

Generative artificial intelligence (AI) has transitioned into a collaborative creator in music composition, raising concerns about its impact on human creative agency. This study investigates how AI automation levels affect creators’ state sense of agency, examining the serial mediating roles of subjective task load and psychological ownership, and the moderating role of musical expertise. A 3 (AI automation level: low, medium, high) × 2 (musical expertise: novice, expert) between-subjects experiment was conducted (N = 162). Participants completed a music co-creation task using standardized AI generative tools. Results revealed that higher AI automation significantly reduced subjective task load, psychological ownership, and state sense of agency. Furthermore, subjective task load and psychological ownership serially mediated the negative relationship between AI automation and state sense of agency. Musical expertise significantly moderated these effects, with experts experiencing a more pronounced decline in psychological ownership and agency under high automation compared to novices. These findings indicate that while high automation reduces creative effort, it alienates creators from their output by diminishing process involvement and belongingness. Future generative AI systems should balance efficiency with agency preservation to support sustainable human-AI co-creation.

… sustaining agency. Preliminary findings reveal that agency in human – AI co-creation fluctuates … Our findings suggest that ownership in human-AI co-creation is not static but shaped by …

… metric, the Creativity Support Index (CSI) that is designed to help researchers and designers evaluate … measurement tool for evaluating creativity support: the Creativity Support Index (…

… a digital creativity support tool to support the creative process of … on concepts from creativity and cognition support tools, which … In concert with other evaluation metrics, the CSI can help …

The design and development of Creativity Support Tools (CSTs) is of growing interest in research at the intersection of creativity and Human-Computer Interaction, and has been identified as a 'grand challenge for HCI'. While creativity research and HCI each have had long-standing discussions about---and rich toolboxes of---evaluation methodologies, the nascent field of CST evaluation has so far received little attention. We contribute a survey of 113 research papers that present and evaluate CSTs, and we offer recommendations for future CST evaluation. We center our discussion around six major points that researchers might consider: 1) Clearly define the goal of the CST; 2) link to theory to further understanding of usage of CSTs; 3) recruit domain experts, if applicable and feasible; 4) consider longitudinal, in-situ studies; 5) distinguish and decide whether to evaluate usability or creativity; and 6) as a community, help develop a toolbox for CST evaluation.

Creativity Support Tools (CSTs) are widely used across diverse creative domains, with generative AI recently increasing the abilities of CSTs. To better understand how the success of CSTs is determined in the literature, we conducted a review of outcome measures used in CST evaluations. Drawing from (n=173) CST evaluations in the ACM Digital Library, we identified the metrics commonly employed to assess user interactions with CSTs. Our findings reveal prevailing trends in current evaluation practices, while exposing underexplored measures that could broaden the scope of future research. Based on these results, we argue for a more holistic approach to evaluating CSTs, encouraging the HCI community to consider not only user experience and the quality of the generated output, but also user-centric aspects such as self-reflection and well-being as critical dimensions of assessment. We also highlight a need for validated measures specifically suited to the evaluation of generative AI in CSTs.

… In summary, the way forward focused on MILCs that used multiple metrics and evaluation techniques based on long-term in-depth observations and interviews over weeks and months …

The design of product-service systems is one of the more recent evolutions in the field of design and innovation. The approach for designing products and services in an integrated way holds the opportunity for developing more value for the user and the entire value chain. Despite the existence of various PSS design tools and methods to optimise this creative development process, it remains unclear to what extent the full array of tools supports the design team in their creative work. In this paper, we present the results of four years of iterative evaluation of a PSS Design Toolkit deployed in a graduate education setting, using the creativity support index (CSI), a psychometricallyvalidated instrument. By using the CSI longitudinally, the results enabled us to iteratively improve the PSS Design Toolkit to better support future generation designers for the challenges that come with designing these product-service systems.

Creativity Support Tools (CSTs) play a fundamental role in the study of creativity in Human-Computer Interaction (HCI). Even so, there is no consensus definition of the term 'CST' in HCI, and in most studies, CSTs have been construed as one-off exploratory prototypes, typically built by the researchers themselves. This makes it difficult to clearly demarcate CST research, but also to compare findings across studies, which impedes advancement in digital creativity as a growing field of research. Based on a literature review of 143 papers from the ACM Digital Library (1999-2018), we contribute a first overview of the key characteristics of CSTs developed by the HCI community. Moreover, we propose a tentative definition of a CST to help strengthen knowledge sharing across CST studies. We end by discussing our study's implications for future HCI research on CSTs and digital creativity.

… Creativity evaluation metrics also reflect this desire for success, often aiming to maximize the number of likes and remixes [31], number of completed projects [28], or scores by expert …

… methodology for evaluating IBI support tools, building on prior creative cognition research in … of ideation metrics of curation. Elemental ideation metrics evaluate creativity within curated …

… a battery of mutually independent ideation metrics to assess creative products [11]. Webb … metrics [18]. Carroll et al. [5] developed a psychometric tool called the Creativity Support Index. …

Creativity Support Tools (CSTs) have become an integral part of artistic creation. The range of CST technologies is broad—from fabricators to generative algorithms to robots. The interaction approaches for CSTs are accordingly broad. CSTs combine specific technologies and interaction types to serve a spectrum of roles and users. In this work, we tackle a comprehensive understanding of how the intersections of users, roles, interactions, and technologies form a design space for CSTs. We accomplish this by reviewing 111 art-creation CSTs from HCI and computing research and analyzing how diverse aspects of CSTs relate to each other. Our findings identify patterns for designing CSTs, which can give guidance to future CST designers. We also highlight under-explored types of CSTs within the HCI community, providing future directions that CST researchers can pursue given the current trajectory of technological advancement. This work contributes an integrating perspective to understand the landscape of art-creation CSTs.

… discovery ideation metrics were applied to assess the creative products. Information representation was shown to significantly impact the emergence and variety ideation metrics (Figure …

… processes, as well as a means for automatically evaluating the results of such … tools. We proceed to describe the Computational Creativity metrics that can be used for evaluating …

… At the same time, the visual interaction style provides an alternative to the dialog-based model employed in most mixed-initiative (MI) systems. Visual thinking tools avoid complex …

The scientific process is inherently creative, requiring the generation and exploration of ideas for scientific inspiration, projects, study design, and communication. As large language models (LLMs) advance rapidly, scientists increasingly take advantage of their abilities. While LLMs show great promise in supporting many steps of the scientific process, researchers still face significant challenges in validating and steering their output. Interactions tailored to scientists and their specific tasks may empower them to harness the full creative potential of LLMs. I present a course of research that will lead to the development and evaluation of mixed-initiative methods for co-creation in scientific research. These methods aim to facilitate verification and control of AI output. I briefly describe my prior and proposed work on mixed-initiative methods for co-creating research inspiration, studies, and communication, and I detail my current project on an LLM-powered tool for co-creating research project ideas.

In this paper, we report on our investigation of human-AI collaboration for mind-mapping. We specifically focus on problem exploration in pre-conceptualization stages of early design. Our approach leverages the notion of query expansion—the process of refining a given search query for improving information retrieval. Assuming a mind-map as a network of nodes, we reformulate its construction process as a sequential interaction workflow wherein a human user and an intelligent agent take turns to add one node to the network at a time. Our contribution is the design, implementation, and evaluation of algorithm that powers the intelligent agent (IA). This paper is an extension of our prior work (Chen et al., 2019, “Mini-Map: Mixed-Initiative Mind-Mapping Via Contextual Query Expansion,” AIAA Scitech 2020 Forum, p. 2347) wherein we developed this algorithm, dubbed Mini-Map, and implemented a web-based workflow enabled by ConceptNet (a large graph-based representation of “commonsense” knowledge). In this paper, we extend our prior work through a comprehensive comparison between human-AI collaboration and human-human collaboration for mind-mapping. We specifically extend our prior work by: (a) expanding on our previous quantitative analysis using established metrics and semantic studies, (b) presenting a new detailed video protocol analysis of the mind-mapping process, and (c) providing design implications for digital mind-mapping tools.

… Mixed-initiative tools are a special case of computer-aided design, where the computer takes on a more proactive role [8, 27]. Mixed-initiative tools rely on both a human initiative and a …

In recent years, there has been a growing application of mixed-initiative co-creative approaches in the creation of video games. The rapid advances in the capabilities of artificial intelligence (AI) systems further propel creative collaboration between humans and computational agents. In this tutorial, we present guidelines for researchers and practitioners to develop game design tools with a high degree of mixed-initiative co-creativity (MI-CCy). We begin by reviewing a selection of current works that will serve as case studies and categorize them by the type of game content they address. We introduce the MI-CCy Quantifier, a framework that can be used by researchers and developers to assess co-creative tools on their level of MI-CCy through a visual scheme of quantifiable criteria scales. We demonstrate the usage of the MI-CCy Quantifier by applying it to the selected works. This analysis enabled us to discern prevalent patterns within these tools, as well as features that contribute to a higher level of MI-CCy. We highlight current gaps in MI-CCy approaches within game design, which we propose as pivotal aspects to tackle in the development of forthcoming approaches.

… being developed as a tool to support creativity through a mixed-initiative composition space. … as ingredients in ideation. combinFormation’s mixed-initiative composition space supports …

Quantitative data is frequently represented using color, yet designing effective color mappings is a challenging task, requiring one to balance perceptual standards with personal color preference. Current design tools either overwhelm novices with complexity or offer limited customization options. We present ColorMaker, a mixed-initiative approach for creating colormaps. ColorMaker combines fluid user interaction with real-time optimization to generate smooth, continuous color ramps. Users specify their loose color preferences while leaving the algorithm to generate precise color sequences, meeting both designer needs and established guidelines. ColorMaker can create new colormaps, including designs accessible for people with color-vision deficiencies, starting from scratch or with only partial input, thus supporting ideation and iterative refinement. We show that our approach can generate designs with similar or superior perceptual characteristics to standard colormaps. A user study demonstrates how designers of varying skill levels can use this tool to create custom, high-quality colormaps. ColorMaker is available at: colormaker.org

… tools that allow humans to foster creativity through collaboration with AI is actively progressing. The concept of an AI-based mixed initiative … the creativity of fashion designers. CoCoStyle …

… Information discovery means ideation in conjunction with information finding. The representation shifts associated with insight and ideation, such as changes in conceptual framing and …

Exploration and Optimization are ubiquitous activities in human work. Exploration allows us to consider multiple design or solution possibilities. Optimization allows us to narrow and refine diverse possibilities. Given that generative AI produces diverse outcomes for the same input – an attribute known as generative variability – there is now a need to re-examine Exploration and Optimization in the context of generative AI and understand their implications for the future of work. The relationship between Exploration and Optimization has traditionally been portrayed as independent, inverse, or even contradictory. Guided by concepts from mixed initiative interfaces, we conducted a conceptual analysis of three future-work-centered generative AI use cases. We show that the relationship between Exploration and Optimization is not “one-size fits all.” We show how Exploration and Optimization may be combined into several integrated and inter-dependent work processes. Our use cases indicate that generative variability will increase the importance of Exploration and Optimization in future work tasks and applications.

Narratives are a predominant part of games, and their design poses challenges when identifying, encoding, interpreting, evaluating, and generating them. One way to address this would be to approach narrative design in a more abstract layer, such as narrative structures. This paper presents Story Designer, a mixed-initiative co-creative narrative structure tool built on top of the Evolutionary Dungeon Designer (EDD) that uses tropes, narrative conventions found across many media types, to design these structures. Story Designer uses tropes as building blocks for narrative designers to compose complete narrative structures by interconnecting them in graph structures called narrative graphs. Our mixed-initiative approach lets designers manually create their narrative graphs and feeds an underlying evolutionary algorithm with those, creating quality-diverse suggestions using MAP-Elites. Suggestions are visually represented for designers to compare and evaluate and can then be incorporated into the design for further manual editions. At the same time, we use the levels designed within EDD as constraints for the narrative structure, intertwining both level design and narrative. We evaluate the impact of these constraints and the system’s adaptability and expressiveness, resulting in a potential tool to create narrative structures combining level design aspects with narrative.

Visual blends are an advanced graphic design technique to seamlessly integrate two objects into one. Existing tools help novices create prototypes of blends, but it is unclear how they would improve them to be higher fidelity. To help novices, we aim to add structure to the iterative improvement process. We introduce a method for improving prototypes that uses secondary design dimensions to explore a structured design space. This method is grounded in the cognitive principles of human visual object recognition. We present VisiFit – a computational design system that uses this method to enable novice graphic designers to improve blends with computationally generated options they can select, adjust, and chain together. Our evaluation shows novices can substantially improve 76% of blends in under 4 minutes. We discuss how the method can be generalized to other blending problems, and how computational tools can support novices by enabling them to explore a structured design space quickly and efficiently.

We investigate how multiple sliders with and without feedforward visualizations influence users’ control of generative models. In an online study (N=138), we collected a dataset of people interacting with a generative adversarial network (StyleGAN2) in an image reconstruction task. We found that more control dimensions (sliders) significantly increase task difficulty and user actions. Visual feedforward partly mitigates this by enabling more goal-directed interaction. However, we found no evidence of faster or more accurate task performance. This indicates a tradeoff between feedforward detail and implied cognitive costs, such as attention. Moreover, we found that visualizations alone are not always sufficient for users to understand individual control dimensions. Our study quantifies fundamental UI design factors and resulting interaction behavior in this context, revealing opportunities for improvement in the UI design for interactive applications of generative models. We close by discussing design directions and further aspects.

As GenAI technologies such as large language models, diffusion models, and multimodal generative systems increasingly permeate design workflows, their implications for creativity, methodology, ethics, and collaboration demand critical scholarly attention. This paper presents a systematic literature review of generative artificial intelligence (GenAI) in user interface (UI) and user experience (UX) design, drawing on fifty peer-reviewed and preprint articles published between 2020 and 2025. The review is structured around five research questions, addressing: (1) the stages of the UI/UX design process where GenAI tools are most actively applied, (2) the methodological approaches used to evaluate their integration, (3) the ethical considerations arising from their use, (4) models of human-AI collaboration in design practice, and (5) the research gaps that shape the future trajectory of this field. Findings indicate that while GenAI tools are widely adopted in prototyping and visual asset generation, their use in early-stage conceptualization and UX evaluation remains limited. The literature also reveals methodological fragmentation and a lack of standardized evaluation frameworks. Ethical concerns surrounding bias, transparency, and privacy are underexplored, and few studies provide robust models for collaborative work between humans and AI. This review identifies the need for longitudinal research, structured participatory frameworks, and ethically grounded design methodologies. The paper contributes a comprehensive synthesis of current knowledge and outlines directions for future inquiry at the intersection of generative AI and human-computer interaction.

In automated UI design generation, a key challenge is the lack of support for iterative processes, as most systems focus solely on end-to-end output. This stems from limited capabilities in interpreting design intent and a lack of transparency for refining intermediate results. To better understand these challenges, we conducted a formative study that identified concrete and actionable requirements for supporting iterative design with Generative Tools. Guided by these findings, we propose PrototypeFlow, a human-centered system for automated UI generation that leverages multi-modal inputs and models. PrototypeFlow takes natural language descriptions and layout preferences as input to generate the high-fidelity UI design. At its core is a theme design module that clarifies implicit design intent through prompt enhancement and orchestrates sub-modules for component-level generation. Designers retain full control over inputs, intermediate results, and final prototypes, enabling flexible and targeted refinement by steering generation and directly editing outputs. Our experiments and user studies confirmed the effectiveness and usefulness of our proposed PrototypeFlow.

The integration of artificial intelligence (AI) into creative workflows has significantly transformed digital media design, enabling more efficient and innovative processes. This study proposes an interactive co-creation framework driven by StyleGAN2-ADA, designed to enhance human-AI collaboration in visual content generation. The purpose of the research is to evaluate the potential of AI-assisted ideation in creative industries, focusing on the development of a real-time system that allows users to manipulate semantic visual attributes such as expression, style intensity, and lighting. A total of 18 participants, including design professionals and students, engaged with the system to perform creative tasks, providing both quantitative and qualitative data on system usability and effectiveness. The study utilized Fréchet Inception Distance (FID), System Usability Scale (SUS), and interaction logs to measure image quality, user satisfaction, and engagement. Key findings include a low FID score of 4.82, an average task completion time of 6.2 minutes, and a SUS score of 84.1, indicating high usability and efficiency. User feedback highlighted the system’s ability to facilitate rapid ideation and foster a productive balance between AI assistance and creative autonomy. The findings suggest that the proposed framework can serve as an effective tool for enhancing the creative process, with potential applications in design, education, and multimedia systems. Future research will explore expanding the framework’s applicability to diverse creative tasks and further improving user interaction features.

… the user performance, workload, and experience for a generative design solution space selection process, … was developed for use in this study to replicate the down select process of the …

Unlike static and rigid user interfaces, generative and malleable user interfaces offer the potential to respond to diverse users’ goals and tasks. However, current approaches primarily rely on generating code, making it difficult for end-users to iteratively tailor the generated interface to their evolving needs. We propose employing task-driven data models—representing the essential information entities, relationships, and data within information tasks—as the foundation for UI generation. We leverage AI to interpret users’ prompts and generate the data models that describe users’ intended tasks, and by mapping the data models with UI specifications, we can create generative user interfaces. End-users can easily modify and extend the interfaces via natural language and direct manipulation, with these interactions translated into changes in the underlying model. The technical evaluation of our approach and user evaluation of the developed system demonstrate the feasibility and effectiveness of the proposed generative and malleable UIs.

… is when a user prompts a generative model via text and/or … a crucial part of the generative process, but these interactions … load and effectively gives the user control over the system. …

… The fact that Petre is primarily concerned with the visual design of knitwear, whereas our study is in the domain of interaction design, may account for some of the differences. Petre et al. …

Inspiration plays an important role in design, yet its specific impact on data visualization design practice remains underexplored. This study investigates how professional visualization designers perceive and use inspiration in their practice. Through semi-structured interviews, we examine their sources of inspiration, the value they place on them, and how they navigate the balance between inspiration and imitation. Our findings reveal that designers draw from a diverse array of sources, including existing visualizations, real-world phenomena, and personal experiences. Participants describe a mix of active and passive inspiration practices, often iterating on sources to create original designs. This research offers insights into the role of inspiration in visualization practice, the need to expand visualization design theory, and the implications for the development of visualization tools that support inspiration and for training future visualization designers.

Throughout the design process, designers encounter diverse stimuli that influence their work. This influence is particularly notable during idea generation processes that are augmented by novel design support tools that assist in inspiration discovery. However, fundamental questions remain regarding why and how interactions afforded by these tools impact design behaviors. This work explores how designers search for inspirational stimuli using an AI-enabled multi-modal search platform, which supports queries by text and non-text-based inputs. Student and professional designers completed a think-aloud design exploration task using this platform to search for stimuli to inspire idea generation. We identify expertise and search modality as factors influencing design exploration, including the frequency and framing of searches, and the evaluation and utility of search results.

Design activity can be hindered by several communication barriers that exist between the designer and the other stakeholders. To overcome these barriers, designers can utilise user-research outputs. In this regard, designers look to user-research results to inspire decisions by providing interpretable outputs, to guide decisions by pointing out possible directions and supporting their arguments and providing justification for their decisions for the persuasion of others. These impacts can be achieved if user research findings are communicated effectively to the designers. In this paper, an interactive information system is introduced that has been developed for the delivery of the results of user research. The core function of the system is to provide inspiration and guidance to the designer, while also assisting them in justifying their decisions. After presenting such a system, the aim is to discuss the requirements for the communication of user research that can be considered while designing future communication media. Positive feedback has been received from the collaborating firm, which used the system, regarding how the system provides guidance through the successful conveyance of multi-dimensional data.

… discourse regarding visual ization aesthetics and interaction design experimentation. This … in graphic design, typography, architecture, interaction design, and visual or media art. …

… inspiration, as they arise through negotiation and transformation, and are mediated by design artefacts during an Inspiration … sources of inspiration, in order to generate design concepts. …

System Engineering education typically includes content to help students learn to design and engineer large, complex systems in a structured way. In this paper, we describe the outcomes of introducing a human-centered design tool, the Inspiration Design Toolkit (IDT), to encourage students to think non-linearly. The IDT is an educational resource consisting of a deck of illustrated cards that contain provocative questions, reflection messages and icons, applicative examples, and key takeaways on microlearning units. The aim of the IDT is to improve the participants’ learning experience and course engagement, increase opportunities for them to interact with their peers and teaching team, enable them to practice and reinforce the concepts through the creation of their own IDT cards, and share the cards in the discussion to increase learners’ engagement with course material and peers. We designed the IDT for an MIT online course on System Thinking. We collected, analyzed, and synthesized qualitative and quantitative feedback from 171 course participants. Our findings suggest that IDT provides learners with a digital asset that allows them to reinforce and recall the course takeaways, and apply them to other contexts. For future research, we want to understand how learners like and use IDT through demographic differences and preferred self-identified learning styles. We discuss how these findings may help educators consider critical design principles and for creating a digital self-learning toolkit connected to the course content and increasing its content adaptability.

Abstract Inspirational stimuli are known to be effective in supporting ideation during early-stage design. However, prior work has predominantly constrained designers to using text-only queries when searching for stimuli, which is not consistent with real-world design behavior where fluidity across modalities (e.g., visual, semantic, etc.) is standard practice. In the current work, we introduce a multi-modal search platform that retrieves inspirational stimuli in the form of 3D-model parts using text, appearance, and function-based search inputs. Computational methods leveraging a deep-learning approach are presented for designing and supporting this platform, which relies on deep-neural networks trained on a large dataset of 3D-model parts. This work further presents the results of a cognitive study (n = 21) where the aforementioned search platform was used to find parts to inspire solutions to a design challenge. Participants engaged with three different search modalities: by keywords, 3D parts, and user-assembled 3D parts in their workspace. When searching by parts that are selected or in their workspace, participants had additional control over the similarity of appearance and function of results relative to the input. The results of this study demonstrate that the modality used impacts search behavior, such as in search frequency, how retrieved search results are engaged with, and how broadly the search space is covered. Specific results link interactions with the interface to search strategies participants may have used during the task. Findings suggest that when searching for inspirational stimuli, desired results can be achieved both by direct search inputs (e.g., by keyword) as well as by more randomly discovered examples, where a specific goal was not defined. Both search processes are found to be important to enable when designing search platforms for inspirational stimuli retrieval.

Designers systematically seek inspiration from various sources from different domains and expand their repertoire of paradigms to foster their designs. However, designers usually do not explore the full potential of their inspiration sources by establishing surface level and close domain analogies regardless of their experience. We think that the process of finding deep and distant domain analogies could be made more understandable for novice designers and less time consuming for professionals with an inspiration/ideation practice for innovation. We discovered that decontextualizing the inspiration source and deconstructing its perceived meaning into fragments lead designers to explore new connections with inspiration source and discover new inspirational points. Here, in this paper, we re-introduce our re-reading in design practice from our previous studies and examine whether it can be a complimentary inspiration/ideation practice for designers to deconstruct conventional paradigms and create deep and distant domain analogies more comfortably.

… to visual, kinaesthetic and verbal aspects of design activity. It was essential to use diverse processing of mental images and multisensory sources of inspiration… social interaction seem to …

Abstract This study investigates a new form of inspirational stimuli deployed in virtual reality (VR) and utilizes them in conceptual design. Compared with prior research that directly embedded 2D inspirational materials in an immersive VR space (immersive 2D), we built completely stereoscopic inspirational stimuli therein (immersive 3D) and allowed designers to generate schemes that could interact with the content of immersive 3D. The impact of expanding the dimensionality was uncovered through a comparative experiment focusing on three aspects: semantic perception, design behaviour, and emotional response. The results showed that immersive 3D stimuli encouraged designers to perceive higher-level abstract semantics, increased the frequency and duration of their observation on inspirational stimuli and affective expression behaviours, and promoted more positive emotions, which all contributed to the novelty and feasibility of design outcomes.

In early stages of creative processes, practitioners externalize and combine inspirational materials, using strategies such as mood board creation to achieve a desired vision and aesthetic. Yet, collecting and combining materials can be difficult: (1) mood boards bias towards 2D images, neglecting audio, video, and 3D models; (2) alternative externalizations such as prototypes are best suited for later stages and can be time-consuming and tedious to create; and (3) online searches lead to disjointed sources between different websites and assets in the file system. To address these challenges, we created MoodCubes, a system for rapid creation and manipulation of multimedia content. When adding content, MoodCubes decomposes objects (e.g., extracting colour palettes), suggests new materials without the need to search (e.g., 3D models, images, lighting effects), and provides filters to change the scene’s aesthetic. We studied eight creative professionals using MoodCubes, which suggested ways the system might advance existing design practices.

… Our goal was to design interaction techniques for visual comparison that resemble these natural … Although the inspiration for our solution comes from natural behavior, we should clearly …

… for human-centered design and UX as key to the successful implementation of GenAI systems, focusing on users… This study underscores the strategic value of UX design in fostering the …

Abstract Generative design uses artificial intelligence-driven algorithms to create and optimize concept variants that meet or exceed performance requirements beyond what is currently possible using the traditional design process. However, current generative design tools lack the integration of human factors, which diminishes the efforts to understand and inject a broad set of human capabilities, limitations, and potential emotional responses for future human-centered product and service innovation. This paper demonstrates collaborative research in formulating a human-centered generative design framework that injects human factors early in the design for quick-and-dirty concept creation and evaluation. Three case studies overviewing our ongoing multidisciplinary research efforts in synthesizing human and mechanical attributes are presented. The results show that the framework has the potential to enhance human factors representation within generative design workflow. Strategies from a computational design perspective, such as data-driven generative design, digital human modeling, and mixed-reality validation, are discussed as alternative approaches that could be implemented to augment designers.

… the-art user experience (UX) frameworks for human-centered AI (HCAI) in generative AI systems. … , integrating holistic user experiences, and emphasizing ethical principles, participatory …

… and cons of Generative AI applications in the UX design process, … the Three Diamond Design Process (Wang et al., 2022). The … Human-centered design thinking is pivotal in the design …

… intended user experience and user interaction with the future system. What follows after the … optimising designs [10]. Different design activities and outcomes exist: Generative design …

Generative AI applications present unique design challenges. As generative AI technologies are increasingly being incorporated into mainstream applications, there is an urgent need for guidance on how to design user experiences that foster effective and safe use. We present six principles for the design of generative AI applications that address unique characteristics of generative AI UX and offer new interpretations and extensions of known issues in the design of AI applications. Each principle is coupled with a set of design strategies for implementing that principle via UX capabilities or through the design process. The principles and strategies were developed through an iterative process involving literature review, feedback from design practitioners, validation against real-world generative AI applications, and incorporation into the design process of two generative AI applications. We anticipate the principles to usefully inform the design of generative AI applications by driving actionable design recommendations.

Attention is a variable and increasingly limited human ability, yet most interactive systems treat it as fixed or infer it implicitly through opaque signals. This PhD research investigates how attention-centered generative user interfaces (GenUIs) can be designed to support diverse attention abilities while preserving user trust and agency. Using reader modes as a design space, the dissertation integrates participatory design, empirical evaluation, and generative prototyping to move beyond content filtering toward user-controlled, ability-based interface generation. Across three research projects, it (1) elicits user expectations and ethical considerations for (attentive) GenUIs through Design Fiction, (2) empirically identifies the limits of contemporary reader modes on comprehension and cognitive load, followed by designs of attention-adaptable generative alternatives, and (3) examines how ability profiles shape perceptions of agency and epistemic trust in GenUIs. The research contributes design knowledge, empirical insights, and inclusive methods for attention-adaptive generative systems in HCI.

… in UX may arise less from the tools themselves than from how AI support reshapes the distribution of design effort within human-centered … and user experience in practical UX contexts. …