Fig 1 - uploaded by Jeba Rezwana
Content may be subject to copyright.
Source publication
Human-AI co-creativity involves both humans and AI collaborating on a shared creative product as partners. In a creative collaboration, interaction dynamics, such as turn-taking, contribution type, and communication, are the driving forces of the co-creative process. Therefore the interaction model is a critical and essential component for effectiv...
Contexts in source publication
Context 1
... understand interaction dynamics in an open-ended improvisational collaboration, Kellas and Trees present a model of interactional sense-making [81]. They describe two types of interaction in the sense-making process: interaction between collaborators and interaction with the shared product ( Figure 1). We adapt and extend this model for COFI to ground our space of possibilities for interaction design on the concept of interactional sense-making. ...
Context 2
... section presents components related to the relational interaction dynamics between the human and the AI as co-creators. As shown in Figure 1(a), interaction between collaborators is divided into two subcategories which are collaboration style and communication style. ...
Similar publications
Purpose of Review
There is increasing evidence that land use and land cover (LULC) change interacts with climate change to shape biodiversity dynamics. The prevailing hypothesis suggests that generalist species have an advantage in novel climatic and land cover conditions, while specialists are expected to be more sensitive to both stressors (gener...
Citations
... Audry (2021) suggests that generative AI forms new kinds of humanmachine relationships by shaping and reconfiguring the agencies involved in creative endeavours. Creativity in text-based generative content has also been approached from the perspective of human-computer co-creation (Rezwana and Maher 2022). Oppenlaender (2022) posits that human creativity in textbased generative art lies not in the final product (digital image) but emerges in the interaction and iterative prompt engineering with AI. ...
The rise of image-generating artificial intelligence (AI) tools has triggered changes in digital art and graphic design, provoking debates in the creative industry. However, scant research exists about children’s and youths’ insights into and encounters with generative AI. Building on sociocultural and new materialist perspectives, this exploratory study proposed to address this gap by exploring middle schoolers’ ( N = 10) creative interaction with generative AI, particularly with text-to-image generative models. Qualitative content analyses of emerging learning activities evidenced how generative AI-formed relations were externalized through novel digital artefacts and collaborative discussions. Ideas evolved through peer collaboration organized around creative making with AI. Teachers facilitated relations between people and technology using dialogic teaching, providing room for unpredictability and critical reflection on the impacts of generative AI, especially authorship and copyright. The study concludes with a discussion of the potential uses of generative AI in future art education research and practice.
... This insight further highlights the need for the careful evaluation and design of co-creative systems. To this end, various frameworks have been developed to support the design of a creative partner [1,50,61], given the challenges of developing effective AI. Moreover, several works have suggested to also logging users' interactions [16,17,42,48] to evaluate their behavior with co-creative AI systems. ...
Crafting a rich and unique environment is crucial for fictional world-building, but can be difficult to achieve since illustrating a world from scratch requires time and significant skill. We investigate the use of recent multi-modal image generation systems to enable users iteratively visualize and modify elements of their fictional world using a combination of text input, sketching, and region-based filling. WorldSmith enables novice world builders to quickly visualize a fictional world with layered edits and hierarchical compositions. Through a formative study (4 participants) and first-use study (13 participants) we demonstrate that WorldSmith offers more expressive interactions with prompt-based models. With this work, we explore how creatives can be empowered to leverage prompt-based generative AI as a tool in their creative process, beyond current "click-once" prompting UI paradigms.
... There are several frameworks that define the possibility space and characteristics of such interactions. One is the Co-Creative Framework for Interaction design (COFI), which breaks down interaction into interaction with the product or between collaborators, and breaks each of those down even further into more sub-categories for understanding interaction [9]. Another framework is Creative Wand, which presents an ontology of three dimensions of communication between the AI agent and human designer (elaborative/reflective, global/local, and human-/agent-initiaed), as well as a modular Python framework for building co-creative systems [4]. ...
Many advancements have been made in procedural content generation for games, and with mixed-initiative co-creativity, have the potential for great benefits to human designers. However, co-creative systems for game generation are typically limited to specific genres, rules, or games, limiting the creativity of the designer. We seek to model games abstractly enough to apply to any genre, focusing on designing game systems and mechanics, and create a controllable, co-creative agent that can collaborate on these designs. We present a model of games using state-machine-like components and resource flows, a set of controllable metrics, a design evaluator simulating playthroughs with these metrics, and an evolutionary design balancer and generator. We find this system to be both able to express a wide range of games and able to be human-controllable for future co-creative applications.
... From early-day research on human interactions with computers [1,32,69], to recent advances in human-AI collaboration [55], initiative has always been studied as one of the key research questions. The most popular discourse on initiative concerns whether direct manipulation or intelligent interface agents should be employed for user interfaces [24,64]. ...
... Research has also proposed a "mixed-initiative" principle to support interactions between humans and computers [32,69]. To support the design of human-AI co-creative systems, Rezwana and Maher proposed the Co-Creative Framework for Interaction design (COFI), which categorized human-AI collaboration styles into spontaneous or planned based on the timing of initiative [55] . They found that most existing co-creative systems support "planned" instead of "spontaneous" timings, suggesting non-improvisational co-creativity. ...
Prewriting is the process of discovering and developing ideas before a first draft, which requires divergent thinking and often implies unstructured strategies such as diagramming, outlining, free-writing, etc. Although large language models (LLMs) have been demonstrated to be useful for a variety of tasks including creative writing, little is known about how users would collaborate with LLMs to support prewriting. The preferred collaborative role and initiative of LLMs during such a creativity process is also unclear. To investigate human-LLM collaboration patterns and dynamics during prewriting, we conducted a three-session qualitative study with 15 participants in two creative tasks: story writing and slogan writing. The findings indicated that during collaborative prewriting, there appears to be a three-stage iterative Human-AI Co-creativity process that includes Ideation, Illumination, and Implementation stages. This collaborative process champions the human in a dominant role, in addition to mixed and shifting levels of initiative that exist between humans and LLMs. This research also reports on collaboration breakdowns that occur during this process, user perceptions of using existing LLMs during Human-AI Co-creativity, and discusses design implications to support this co-creativity process.
... In this short paper, we provide several examples of one answer to that question. Re-using the Collective concept from the 5Cs of Kantosalo and Takala (2020), we propose that one type of creativity may emerge asymmetrically in the interactional (Rezwana and Maher 2022) spaces between human and AI. We illustrate this conjecture through human-AI dialogs using a highly-interactive UI, intermediating between human actions and the GPT-3-turbo large language model (OpenAI 2023). ...
... We note that the metaphorical design could not have been created by the UI alone. We would like to claim that the co-creativity occurred interactionally between human and AI, similar to the concept of Rezwana and Maher (Rezwana and Maher 2022), but in a more conversational environment. ...
The locus of co-creativity in human-AI creative tasks has not been resolved. We explore a sub-part of this problem through the use of analogies for reframing a conceptual design task. In our scenarios, the human proposes an analogy, and then the human and a conversational UI to a Large Language Model (LLM) collab-oratively explore design features based on that analogy. In one scenario, the human asks the AI to propose the analogy. In our experiments, co-creativity occurs in the interactional shared space between human and AI.
... Taking the AI-TAM as a point of departure, we added the process variables of user control, AI output transparency, perceived partnership and replaced the output variables collaborative intention and behavioral intention with willingness to train, willingness to co-develop and willingness to adopt. These variables were inspired by fundamental principles of HI (mutual learning), HCAI (pursuit of high levels of automation and control simultaneously), and IML (continuous interactivity) as well as the Co-Creative Framework for Interaction Design (COFI) which emphasizes the importance of establishing a partnership between end users and AI support tool [27]. The links represented in Figure 1 are supported by exploratory data analyses described in Section 4.1 and any omitted variables should be validated in further studies. ...
... In an effort to combat algorithmic overconfidence, future designs could display GD assistant's level of confidence in the generated design as a means of the algorithm communicating to the end user. Bi-directional communication has also been shown to increase perceived partnership [27]. ...
... Perceived partnership is one of the new constructs we added to the HI-TAM given its importance in human-AI co-creative systems [27]. Our HI-TAM shows quantitative correlations and qualitative co-occurrences between perceived partnership and perceived usefulness. ...
The emergence of generative design (GD) has introduced a new paradigm for co-creation between human experts and AI systems. Empirical findings have shown promising outcomes such as augmented human cognition and highly creative design products. Barriers still remain that prevent individuals from perceiving and adopting AI, entering into collaboration with AI and sustaining it over time. It is even more challenging for creative design industries to adopt and trust AI where these professionals value individual style and expression, and therefore require highly personalized and specialized AI assistance. In this paper, we present a holistic hybrid intelligence (HI) approach for individual experts to train and personalize their GD assistants on the fly. Our contribution to human-AI interaction is three-fold including i) a programmable common language between human and AI to represent the expert’s design goals to the generative algorithm, ii) a human-centered continual training loop to seamlessly integrate AI-training into the expert’s task workflow, iii) a hybrid intelligence narrative to address the psychological willingness to spend time and effort training such a virtual assistant. This integral approach enables individuals to directly communicate design goals to AI and seeks to create a psychologically safe space for adopting, training and improving AI without the fear of job-replacement. We concertize these constructs through a newly developed Hybrid Intelligence Technology Acceptance Model (HI-TAM). We used mixed methods to empirically evaluate this approach through the lens of HI-TAM with 8 architectural professionals working individually with a GD assistant to co-create floor plan layouts of office buildings. We believe that the proposed approach enables individual professionals, even non-technical ones, to adopt and trust AI-enhanced co-creative tools.
... Gero and Chilton discussed the issue of ownership in human-AI co-creation as some participants felt less ownership over the final co-created product while working with a co-creative system [22]. Additionally, there has been discussion about whether humans or AI should lead the creative process in a human-AI co-creation [53,55] and if AI should assist or collaborate with users [67]. Buschek et al. demonstrated that accountability is one of the major pitfalls when designing co-creative systems [11]. ...
... Like that work we exclude socalled "fire-and-forget" systems (Ramamurthy et al. 2022) from consideration and focus on systems that allow iterative improvement on the creative artifact. A co-creative agent (Rezwana and Maher 2021;2022;Guzdial and Riedl 2019;Grabe 2022;Kreminski and Mateas 2021) is one that AI "possesses the ability to alter the creative work equal to a human counterpart". Note that when it comes to capability, "ability" does not imply human-parity. ...
... To study the information that flows between the user and the MI-CC agent, we depart from the definition of Communications from (Lin, Agarwal, and Riedl 2022), which is itself based on works in categorizing or differentiating between different types of interactions between parties in an MI-CC system. The framework by Rezwana and Maher (2021;2022) that models interactions in co-creative systems that "focus on flow of information between collaborators", inspiring the definition of Communications. Guzdial and Riedl (2019) point out that human designers and AI "can initiate the same action sets to modify the creative work, albeit with different executions". ...
Generative Artificial Intelligence systems have been developed for image, code, story, and game generation with the goal of facilitating human creativity. Recent work on neural generative systems has emphasized one particular means of interacting with AI systems: the user provides a specification, usually in the form of prompts, and the AI system generates the content. However, there are other configurations of human and AI coordination, such as co-creativity (CC) in which both human and AI systems can contribute to content creation, and mixed-initiative (MI) in which both human and AI systems can initiate content changes. In this paper, we define a hypothetical human-AI configuration design space consisting of different means for humans and AI systems to communicate creative intent to each other. We conduct a human participant study with 185 participants to understand how users want to interact with differently configured MI-CC systems. We find out that MI-CC systems with more extensive coverage of the design space are rated higher or on par on a variety of creative and goal-completion metrics, demonstrating that wider coverage of the design space can improve user experience and achievement when using the system; Preference varies greatly between expertise groups, suggesting the development of adaptive, personalized MI-CC systems; Participants identified new design space dimensions including scrutability -- the ability to poke and prod at models -- and explainability.
... Creative AI systems featured highly in this cluster. Creative AI also known as computational creativity is defined as a subfield of AI research that builds computational systems that produce collaborative creative artefacts such as music, artwork, games, literature and design (Rezwana & Maher, 2022;Guzdial & Reidl, 2019;Colton & Wiggins, 2012) that offer learnings for the creation of applications for information systems. This positions the creative AI as an emergent intelligent creative partner that shares a joint creative goal with humans (Guzdial & Reidl, 2019) and is consistent with the findings on the importance of collaborative, adaptable and self-learning bi-lateral systems of humanmachine symbiosis (Charnley et al., 2012;Cook et al., 2019;). ...
The augmentation of human capability with artificial intelligence is integral to the advancement of next generation human-machine collaboration technologies designed to drive performance improvement and innovation. Yet we have limited understanding of how organisations can translate this potential into creating sustainable business value. We conduct an in-depth literature review of interdisciplinary research on the challenges and opportunities in organisational adoption of human-AI collaboration for value creation. We identify five research positions central to how organisations can integrate and align the socio-technical challenges of augmented collaboration, namely strategic positioning, human engagement, organisational evolution, technology development and intelligence building. We synthesise the findings into an integrated model that focuses organisations on building the requisite internal microfoundations for the systematic management of augmented systems.
... A human-LM co-writing system involves both human and LM collaborating on a shared writing artifact (e.g., essay, story, paper, arXiv:2303.06333v1 [cs.HC] 11 Mar 2023 etc.) as partners [10]. Grounded on the Co-Creative Framework for Interaction Design (COFI) model [10], we identify three key components in co-writing systems, including human, LM, and writing artifact. ...
... [cs.HC] 11 Mar 2023 etc.) as partners [10]. Grounded on the Co-Creative Framework for Interaction Design (COFI) model [10], we identify three key components in co-writing systems, including human, LM, and writing artifact. They act as the joint driving force of the interactive co-writing process. ...
A surge of advances in language models (LMs) has led to significant interest in using LMs to build co-writing systems, in which humans and LMs interactively contribute to a shared writing artifact. However, there is a lack of studies assessing co-writing systems in interactive settings. We propose a human-centered evaluation framework, Parachute, for interactive co-writing systems. Parachute showcases an integrative view of interaction evaluation, where each evaluation aspect consists of categorized practical metrics. Furthermore, we present Parachute with a use case to demonstrate how to evaluate and compare co-writing systems using Parachute.
