Integrating Chatbots in Creative Design Learning

 

Lakshay Bareja ¹, Subhashini R ², Sangeet Saroha ³, Sudha Rani ⁴, Chaitanya Joshi ⁵, Sonia Arora ⁶, Bipin Sule ⁷

 

¹ Centre of Research Impact and Outcome, Chitkara University, Rajpura- 140417, Punjab, India

² Assistant Professor, Department of Computer Science and Engineering, Presidency University, Bangalore, Karnataka, India

³ Greater Noida, Uttar Pradesh 201306, India

⁴ Assistant Professor School of Sciences, Noida International, University, India 203201           

⁵ Assistant Professor, Department of Film and Television, Parul Institute of Design, Parul University, Vadodara, Gujarat, India

⁶ Assistant Professor, Department of Computer Science and Engineering (AIML), Noida Institute of Engineering and Technology, Greater Noida, Uttar Pradesh, India

⁷ Department of Development of Enterprise and Service Hubs, Vishwakarma Institute of Technology, Pune, Maharashtra, 411037, India

 

 

1. INTRODUCTION

The fast digital revolution of the creative education has reimagined how students learn to interact with design, visual communication, and trial and error in a studio setting Ramandanis, D., and Xinogalos, S. (2023), Varitimiadis et al. (2021).  Conventional pedagogy of design, historically rooted in the mentoring and physical discovery of people, now has a digital counterpart that is capable of providing multimodal instruction, adaptive learning courses, as well as real-time feedback looping Siddique, S., and Chow, J. C. L. (2020).  With the ongoing development of artificial intelligence, AI-based learning tools, especially chatbots, have become influential facilitators of creative problem-solving, offering round-the-clock support, contextual brainstorming, design feedback, and knowledge search in accordance with the specific needs of learners Liebrecht, C., and Van Hooijdonk, C. (2020) , Siddique, S., and Chow, J. C. L. (2021).  The tools fill gaps in access and minimize the cognitive load in a complex design task, and facilitate exploratory thinking through the generation of alternative ideas and visual directions that are otherwise difficult to imagine at early design phases Kovacek, D., and Chow, J. C. L. (2021) , Rebelo et al. (2022). In this changing environment, there is now a great deal of demand to provide intelligent support in the design ideation and iteration tasks, which are particularly the more students have to deliver conceptually rich, visually coherent, technically sound work under time pressure Xu et al. (2021), Engeness et al. (2025). Chatbots also provide a dynamic layer of conversation, which leads to reflective thinking, the scaffolding of creativity, and expert-like feedback on design Gill et al. (2023). Lund et al. (2023). Nevertheless, even though it sounds promising, there are still important issues related to biasness in the suggestions made by AI, excessive reliance on automated feedback, and the fact that it remains quite hard to implement such systems into pre-existing models of pedagogy Davare et al. (2025).  Chow et al. (2023).

It is these gaps that have inspired this study to examine how chatbot-based systems may be systematically integrated into creative design learning to support ideation fluency, visual reasoning and learner engagement Gosak et al. (2024). The research will (i) examine student-AI-conversational tools, (ii) assess how it influences their idea generation, speed of iteration, and clarity, and (iii) suggest a strong pedagogical system of human and AI-based co-creation in design learning Pandey, S., and Sharma, S. (2023), Chien et al. (2022). The contributions also entail the creation of a structured framework to chatbot-assisted feedback, a multiple-stage workflow to hone the design, and empirical checks on the user studies of design learners and educators Chiu et al. (2023), Chuang et al. (2023).

 

2. Literature Review

Chatbot and artificial intelligence technologies have progressively become part of the modern educational ecosystem and provide scaffolding applications in discipline-specific learning, such as personalized guidance, dynamic assessment, and interactive learning experiences Xu et al. (2021). Chatbots in education are cognitive extensions of learners which assist in real time resolution of inquiries, reinforcement of concepts and reflective conversations, thus lessening the reliance on instructor availability. Research shows that conversational agents based on AI increase engagement, scaffolding, and continuous learning, especially when working with tasks that require, through refinement and exploration of concepts, over and over Engeness et al. (2025). Their application in learning management systems and design studios indicates how chatbots can not only be used as an information search tool but also as a smart assistant that can imitate critique, ideation-inducing, and peer-like engagement in a creative environment Engeness et al. (2025). The innovative design education has developed along the path of the traditional, more studio-based, apprenticeship-based learning to the integrative and even entirely digital forms that welcome the technology-based experimentation. In the past, design education was focused on face-to-face criticism, cycles of ideation, the exploration of materials, and embodied learning as a result of practice-based approaches. As blended learning emerged, the design programs started adopting digital sketching, interactive features, online critique, and preserving the major principles of creativity, aesthetics, and user-oriented thinking Engeness et al. (2025). With the introduction of digital methods with the assistance of cloud computing, simulations, multimodal interfaces, and AI-based assistants, exploratory design has become more and more open, and learners have the ability to visualize alternatives, restructure ideas, and get organised feedback outside of the confines of the classroom. These pedagogical changes represent a continuing trend of shifting to hybrid ecosystems in which human creativity is enriched with intelligent systems that supplement the process and results of design learning Engeness et al. (2025).

In line with the development of the pedagogy, the patterns of human-AI collaboration become prominent in the process of art and design. Generative models, multimodal learning systems, and design-support agents now facilitate brainstorms, reference synthesis, manipulation of visual components and/or design coherence. It is proposed that AI will become a co-creator, and this will provide alternative stylistic directions, pattern variations, and conceptual reinterpretations that may provoke divergent thinking Engeness et al. (2025). The frameworks of human-AI collaboration do not make AI a substitute of creative judgment but a starting point of more intensive ideation so that learners cover a wider conceptual field. The emphasis of these models to collaborative intelligence, in which the designers can retain control over the aesthetic decisions but exploiting machine intelligence to enable quicker iteration, making errors, and synthesizing explorations Engeness et al. (2025).  These associations strengthen the purpose of AI as an enabler of creativity but not an instigator of artistic work.

In spite of these developments, the current studies have a number of gaps in the field of generative feedback and creative cognition in design learning. Most of the studies are devoted to the usability of chatbots or superficial interactions, little is paid to the impact of AI-generated feedback on cognitive processing, design thinking, and long-term skill acquisition Engeness et al. (2025). Generative models have the potential to be used to provide a variety of suggestions; however, the pedagogical implications of using generative models have not been examined thoroughly, especially regarding how students can interpret, trust, or criticize automated responses. Also, there is limited empirical data about the effects of AI-based conversational tools on the iterative design process, improvement, and reflective practice. Such issues as bias, hallucination, and contextuality of AI productions also demonstrate the necessity of sound schemes that would guarantee the responsible implementation of AI in creative education Engeness et al. (2025). These gaps need to be addressed in order to come up with the design of chatbot systems that do not meanlessly supplement human creativity, promote independent thinking, and support pedagogical rigor in the context of design learning.

Table 1

 

3. Conceptual Framework for Chatbot-Integrated Creative Learning

3.1. Role of chatbots as design mentors, collaborators, and evaluators

Chatbots incorporated into creative design education have a dual role of a mentor, colleague, and critic and assist in various stages of a design process. In their role as design mentors, they help learners to refine ideas, comprehend design concepts, and experiment with the visual direction by providing contextualized suggestions and step-by-step instructions. The fact that they can give formative feedback on a regular basis will enable them to spot areas of deficiency in their understanding, evaluate compositions, and rehearse in a more productive way. Chatbots can be used as collaborators to create moodboards, color palettes, layout variations, and other elements of co-creation, as well as to encourage divergent thinking by providing an individual with creative prompts. Such a partnership enables the learners to expand their conceptual search capacity and do anything freely without the fear of being judged. Chatbots, as evaluators, can be used to evaluate the design coherence, functional fit to user requirements, and aesthetic harmony by comparing the student work to learned patterns, design intuition or on rubric-based criteria. The triadic role mentor, collaborator, evaluator is the role described as chatbots as dynamic creative partners, which helps one to gain greater autonomy, less cognitive load, and improve it through the iterative mechanism in a design learning process. The Figure 1 represents an AI-enriched learning environment in which students are exposed to design tools, databases of content, chatbot interfaces, and evaluation tools. The built-in LMS concentrates resources, which allows offering customized advice, automatic feedback, and enhancing creative skill formation due to continuous AI-based assistance and adaptive learning processes.

Figure 1

Figure 1 AI-Integrated Learning Ecosystem for Creative Design and Visual Storytelling

 

3.2. Cognitive and affective dimensions of learner–chatbot interaction

The learner chatbot interaction involves both cognitive and affective aspects that have a considerable impact on creative learning results. At the cognitive level, chatbots facilitate the stimulation of ideas, facilitate the organization of problem-solving efforts, and facilitate information overload through concise and context-infused assistance in complicated design tasks. This cognitive enhancement reinforces formation of ideas, visual reasoning and reflection by encouraging the learners to express ideas, give reasons and seek alternative solutions. Affective aspects are also to be taken into consideration: chatbots allow building a low-stress environment when learners can feel safe to experiment, ask questions, and commit mistakes without the threat of critical assessment. Their tone of conversation can serve to increase motivation, interest, and confidence particularly among those students that might be reluctant to request the help of an instructor. Sense of psychological safety supported by emotional encouragement, a sense of timely encouragement, and understanding responses help in the exploration of more creative possibilities. A combination of these mental and emotional relations develops a harmonized ecosystem that allows creativity to thrive both in its intellectual and emotional support.

 

3.3. Theoretical background Co-creation, constructivism and multimodal learning

The theoretical foundation of the chatbot-based creative learning idea is the constructivist theory, according to which the active role of learners in knowledge building is based on the interaction, experiment, and reflection. Chatbots are constructivist based, as they encourage inquiry, help refine and even encourage learners to construct meaning through dialogues. The theory of co-creation also reinforces the paradigm by making design a collaborative process where the ideas are formed through mutual input of both humans and AI. In this respect, chatbots serve as innovative collaborators that provoke new thinking, add more value to the conceptual, and trigger innovation through a two-way conversation. Multimodal learning theory is another theory that forms a further base on the significance of visual, textual, and interactive modalities in designing education. Chatbots with the ability to produce diagrams, color proposals, sketches and semantic descriptions allow learners to combine more than one mode of representation, and improve conceptualization and creative synthesis. Constructivism, co-creation, and multimodal learning are three inseparable components of a unified theoretical framework, which allows creating a rich, interactive, and mentally stimulating creative learning environment enhanced with intelligent chatbot systems.

 

4. Functional Components and System Architecture

4.1. Dialogue Control and Contextualization

The essence of chatbot-based creative learning systems is dialogue management and contextual understanding to create interactive and meaning-ful and design-relevant interactions. The dialogue manager coordinates user-AI interaction by making sense of users queries, continuing the conversation, and choosing system responses accordingly by intent, context, and design task goals. Elaborate natural language understanding (NLU) models read linguistic forms, semantic hints as well as design-based vocabulary to uncover the desires of the user, whether ideation support, critique seeking, refinement direction or conceptual elucidation. Tracking of the context is imperative because creative work processes are characterized by exploration; the system should be able to recall past design options, user preferences, the history of the feedbacks and current project constraints. This involves the ability to sustain contextual states over several turns of conversation, make references to past conversations and dynamically modify responses. The architecture generally incorporates transformer models with the ability to do deep contextual encodings, multimodal encoders to read visual information and decision-making layers to decide which kind of assistance is most pedagogically useful.

 

4.2. Design-Knowledge Retrieval and Generative Suggestion Engine  

The chatbot system relies on the design-knowledge retrieval and generative suggestion engine as the intellectual support that helps it to deliver appropriate insights, stylistic suggestions, and conceptual variants in the creative process. Knowledge retrieval is the indexing of the structured and unstructured repositories including design principles, case studies, visual examples, typographic rule references, reference color theory, and domain specific heuristics and matching them with a user query using a semantic search and embedding based similarity algorithm. This enables the system to provide contextually relevant knowledge in accordance to the task at hand of a learner. In addition to the retrieval, the generative suggestion engine employs powerful generative model (i.e., diffusion models, GANs, text generators based on transformers, etc.) to suggest new layouts, color palettes, visual treatments, or conceptual directions. Such generative outputs provoke original thinking and increase the search space of the creative user.

 

4.3. Visual and Textual Feedback Modules

The visual and textual feedback modules offer multimodal feedback to learners which is important in the effective creative development process. The textual feedback feature interprets user postings like description of design, statement of concept, or posted objects and presents systematic analysis of evaluation executing on clarity, coherence, composition, use of color, typographic harmony and compatibility with design principles. Based on the rubric-based models and the rule-based evaluators, the chatbot is able to identify strengths, shedding light on weaknesses, and provide steps of improvement that can be implemented. Such descriptions do not only perfect the artifact but enhance conceptual knowledge like a deeper insight. The visual feedback system builds on this feature by using computer vision algorithms, such as feature detection, segmentation, saliency model, and aesthetic scoring models, to interpret images, sketches, layouts or prototypes.

 

4.4. Safety, Bias Minimization, and Transparency Mechanisms

The mechanisms of safety, bias reduction, and transparency, form the necessary components of the responsible and ethical integration of chatbots into the creative learning settings. The safety layer is a safety measure that oversees interactions to avoid malicious, misconstruing, or unsuitable content by abusing filters, toxicity classifiers, and hallucination-detecting algorithms. Such precautions are necessary to make sure that the answers of the chatbot will be pedagogically correct and culturally appropriate. Creative design is a specific area where prejudice reduction is especially important because aesthetics norms, cultural themes, and stylistic allusions might unwillingly mirror biases in training data. To solve this, the system has used fairness-conscious models, equal datasets, and counterfactual assessment to provide a fair representation of feedback and generative suggestions.

Table 2  

 

5. Methodology

5.1. User Study Design Involving Students, Educators, and Professionals  

The user study would represent a wide variety of opinions since it would include undergraduate design students, graduate learners, profession designers, and experienced educators. The participants were split into the controlled groups, in which one side used the traditional design processes whereas the experimental group used the workflows supported by chatbots.

Every participant was subjected to a set creative task, which was followed by reflection and assessment questionnaires. Teachers were used to give rubric-based evaluations, professionals to give qualitative information in terms of design consistency, novelty, and quality of refinement. The data gathered through observational methods, interaction transcript and user feedback were applied to comprehend how the integration of chatbot affected learning behaviors, creativity, clarity of communication and decision making patterns. It was a multi-stakeholder method that guaranteed a comprehensive assessment of the pedagogical and practical effectiveness of the system. In Figure 2, the evaluation process is organized in the form of a participant group, task implementation, reflection, survey, and professional assessment. The framework merges both qualitative and quantitative data to identify both the pedagogical and practical effectiveness of creative design workflows with the assistance of a chatbot.

Figure 2

Figure 2 Evaluation Framework for Comparing Traditional and Chatbot-Assisted Design Workflows

 

Every participant was subjected to a set creative task, which was followed by reflection and assessment questionnaires. Teachers were used to give rubric-based evaluations, professionals to give qualitative information in terms of design consistency, novelty, and quality of refinement. The data gathered through observational methods, interaction transcript and user feedback were applied to comprehend how the integration of chatbot affected learning behaviors, creativity, clarity of communication and decision making patterns. It was a multi-stakeholder method that guaranteed a comprehensive assessment of the pedagogical and practical effectiveness of the system. In Figure 2, the evaluation process is organized in the form of a participant group, task implementation, reflection, survey, and professional assessment. The framework merges both qualitative and quantitative data to identify both the pedagogical and practical effectiveness of creative design workflows with the assistance of a chatbot.

 

5.2. Experimental Organization and Performance

The experimental condition was set in such a way that a realistic creative learning setting was recreated in which participants were asked to perform design tasks in a traditional workflow and a chatbot-enhanced workflow. The system was installed on a secure cloud system with multimodal input features, users could send sketches, visual arrangements and written descriptions to the chatbot interface. The participants were then offered a short orientation module which is a description of what the chatbot could do, what it could not do, and how to communicate. Each session involved a design brief, which set the goals of tasks, limitations and assessment criteria. After that they would undergo three stages; preliminary ideation, refinement of the mid-task, and completion of the design.

 

6. Results and Discussion

6.1. Influence on Ideation Cue and Creativity Improvement

Chatbots have also positively impacted the ideation fluency of students as they have allowed them to quickly test out alternative ideas, styles and visual paths. The participants said that the chatbot was able to help them overcome their creativity block since it proposed a variety of prompts, and thematic variations, as well as conceptual frameworks that made them think more broadly. The cyclic dialogical process aided the divergent ideation process by promoting the refinement, re-framing, or elaboration of ideas at a faster rate than the traditional workflow did. The improvement of creativity was also shown in the depth of the concepts presented, the innovativeness of design solutions, and the enhanced skills to explain the logic of decisions. The judgment of experts revealed that students receiving the support of chatbots created more original pieces of work, stylistically coherent, and conceptually stratified.

 

6.2. Visual Reasoning and Conceptual Clarity

The statistical results are clearly shown that chatbot-supported workflows had a significant positive impact on the visual reasoning and conceptual clarity of learners in various aspects. The accuracy of the visual hierarchy rose to 84 per cent, as opposed to 68 per cent, which means that learners were able to comprehend the way to organize and prioritize graphical items more effectively with the help of multilayered feedback provided by the chatbot. On the same note, the consistency metric of color harmony scale increased by 34.3 to indicate that the AI-generated palette recommendations and contrast ratings facilitated the skill of learners in creating harmonious and appealing color schemes. The balance score of layouts had increased tremendously, which is indicative of the chatbot success in noticeable discrepancies of spaces and providing remedial changes.

Table 3 

Table 3 Comparative Evaluation of Traditional vs. Chatbot-Assisted Design Workflows across Key Visual Communication Metrics
Metric	Traditional Workflow (%)	Chatbot-Assisted Workflow (%)	Improvement (%)
Visual Hierarchy Accuracy	68	84	+23.5
Color Harmony Consistency	64	86	+34.3
Layout Balance Score	61	82	+34.4
Concept Explanation Clarity	58	88	+51.7
Design Rationale Articulation	63	89	+41.3
Error Identification Accuracy	56	83	+48.2

 

The conceptual comprehensiveness also enhanced: the clearness of concept explanation and articulation of design exhibited the most improvement 51.7% and 41.3% respectively. It implies that conversational prompting made learners ponder their choices, explain, and defend the choice of design options more accurately.

Accuracy in the error identification improved to 83% as compared to a previous percentage of 56% thus revealing that chatbot critique and annotation mechanisms enhanced the analytical ability of learners. Altogether, the findings emphasize the idea that chatbots can not only improve visual quality of outputs but also make a valuable contribution to more profound conceptual knowledge and reflective design thinking, which makes them useful in pedagogic work in creative education. Figure 3 demonstrates that the workflows assisted by the chatbot perform better than the conventional approaches in all design aspects such as hierarchy accuracy, color compatibility, clarity, and error detection. The stable positive trend is the indication of the high ability of AI to increase the level of design quality, accuracy, and communication efficiency.

 

Figure 3

Figure 3 Workflow comparison between Traditional

 

Accuracy in the error identification improved to 83% as compared to a previous percentage of 56% thus revealing that chatbot critique and annotation mechanisms enhanced the analytical ability of learners. Altogether, the findings emphasize the idea that chatbots can not only improve visual quality of outputs but also make a valuable contribution to more profound conceptual knowledge and reflective design thinking, which makes them useful in pedagogic work in creative education. Figure 3 demonstrates that the workflows assisted by the chatbot perform better than the conventional approaches in all design aspects such as hierarchy accuracy, color compatibility, clarity, and error detection. The stable positive trend is the indication of the high ability of AI to increase the level of design quality, accuracy, and communication efficiency.

 

6.3. Case Studies of Chatbot-Supported Projects of Design

Case studies showed various advantages of chatbot implementation in various types of projects such as branding, poster designing, user interface/user experience interface, and conceptualizing. According to the branding projects, students utilized the chatbot to produce several stylistic variations as well as to improve brand stories and create more coherent visual identities. Regarding poster design activities, the visual evaluation module of the chatbot helped learners improve the typographic hierarchy and layout structure with the help of the iterative feedback system. The chatbot was useful in the UI/UX projects to evaluate the usability patterns, propose interface elements, and criticize the navigation.

Table 4 

Table 4 Challenges Analysis Table with Numeric Evaluation
Parameter	Risk Severity (1–10)	Observed Frequency (%)	Impact on Learning (%)	Mitigation Effectiveness (%)	Overall Criticality (%)
Over-Reliance on AI	7	42	38	55	56
Bias in Suggestions	6	28	31	61	45
Hallucinated Feedback	5	22	27	58	41
Inconsistent Accuracy	8	35	43	52	59
Cultural/Aesthetic Misalignment	6	19	34	67	39
Misinterpretation of Queries	7	31	36	60	52

 

Figure 4 demonstrates that the main risks of AI-assisted workflows are hallucinated feedback, cultural misfit, and query misinterpretation. Mitigation effectiveness is high although the levels of severity are moderate-to-high. The chart emphasizes the significance of constant monitoring that is aimed at providing reliable, culturally conscious, and pedagogically safe AI-inspired creative spaces.

Figure 4

Figure 4 Risk Severity, Frequency, Impact, and Mitigation Effectiveness in AI-Assisted Creative Workflows

 

7. Conclusion

This paper has shown, through a creative design learning context, that chatbot implementation has a meaningful contribution to ideation fluency, visual reasoning, and engagement by acting as an adaptive mentors, a collaborator, and an evaluator. According to the results, creativity scores, conceptual clarity, layout balance, and color harmony significantly improved as a result of multimodal feedback on the chatbot, their generative guessing opportunity, and tailored advice. Learners were also able to enhance their reflective thinking, autonomy, and efficient iteration processes through the ongoing and contextual dialogue and real-time critique. The case studies also established that chatbot-assisted processes can be successfully applied in various design projects (branding and poster designing, UI/UX and conceptual illustration) indicating the system as versatile and pedagogically useful. This impact was confirmed by the methodology in the categories of students, educators and professionals by means of quantitative measurement, qualitative assessment and analysis of user experiences. Nevertheless, other issues found in the research are risks of over-reliance, inconsistent accuracy, bias in generative results and sometimes hallucinated feedback. The responsibility of ethics in terms of authorship, data privacy, and cultural sensitivity shows that responsibility in execution frameworks is necessary. Regardless of these shortcomings, the results confirm that chatbots have the potential to complement considerably the usual and the blended studio learning when implemented with transparency, human control, and effective safety measures. The research places chatbots in the central role of transformational aids which assist in co-creation, multimodal exploration, and reflective practice of design. Closing the feedback accessibility and encouraging more intensive thinking opportunities, chatbot-enhanced systems represent a valuable direction in the more inclusive, flexible, and forward-thinking creative design education.

 

CONFLICT OF INTERESTS

None. 

 

ACKNOWLEDGMENTS

None.

 

REFERENCES

Chien, Y.-Y., Wu, T.-Y., Lai, C.-C., and Huang, Y.-M. (2022). Investigation of the Influence of Artificial Intelligence Markup Language-Based LINE ChatBot in Contextual English Learning. Frontiers in Psychology, 13, 785752. https://doi.org/10.3389/fpsyg.2022.785752

Chiu, T. K. F., Moorhouse, B. L., Chai, C. S., and Ismailov, M. (2023). Teacher Support and Student Motivation to Learn with Artificial Intelligence (AI) Based Chatbot. Interactive Learning Environments, 31, 1–17. https://doi.org/10.1080/10494820.2023.2238712

Chow, J. C. L., Sanders, L., and Li, K. (2023). Design of an Educational Chatbot Using Artificial Intelligence in Radiotherapy. AI, 4, 319–332. https://doi.org/10.3390/ai4010015

Chuang, C.-H., Lo, J.-H., and Wu, Y.-K. (2023). Integrating Chatbot and Augmented Reality Technology into Biology Learning during COVID-19. Electronics, 12, 222. https://doi.org/10.3390/electronics12010222

Davar, N. F., Dewan, M. A. A., and Zhang, X. (2025). AI Chatbots in Education: Challenges and Opportunities. Information, 16, 235. https://doi.org/10.3390/info16030235

Engeness, I., Nohr, M., and Fossland, T. (2025). Investigating AI Chatbots’ Role in Online Learning and Digital Agency Development. Education Sciences, 15, 674. https://doi.org/10.3390/educsci15060674

Gill, S. S., Xu, M., Patros, P., Wu, H., Kaur, R., Kaur, K., Fuller, S., Singh, M., Arora, P., Parlikad, A. K., et al. (2023). Transformative Effects of ChatGPT on Modern Education: Emerging Era of AI Chatbots. Internet of Things and Cyber-Physical Systems, 4, 19–23. https://doi.org/10.1016/j.itcps.2023.03.002

Gosak, L., Pruinelli, L., Topaz, M., and Stiglic, G. (2024). The ChatGPT Effect and Transforming Nursing Education with Generative AI: Discussion Paper. Nurse Education in Practice, 75, 103888. https://doi.org/10.1016/j.nepr.2024.103888

Kovacek, D., and Chow, J. C. L. (2021). An AI-Assisted Chatbot for Radiation Safety Education in Radiotherapy. IOP SciNotes, 2, 034002. https://doi.org/10.1088/2632-2153/abf447

Liebrecht, C., and van Hooijdonk, C. (2020). Creating Humanlike Chatbots: What Chatbot Developers Could Learn from Webcare Employees in Adopting a Conversational Human Voice. In International Workshop on Chatbot Research and Design (pp. 51–64). Springer. https://doi.org/10.1007/978-3-030-50522-3_5

Lund, B. D., Wang, T., Mannuru, N. R., Nie, B., Shimray, S., and Wang, Z. (2023). ChatGPT and a New Academic Reality: AI-Written Research Papers and the Ethics of the Large Language Models in Scholarly Publishing. Journal of the Association for Information Science and Technology, 74, 570–581. https://doi.org/10.1002/asi.24750

Pandey, S., and Sharma, S. (2023). A Comparative Study of Retrieval-Based and Generative-Based Chatbots Using Deep Learning and Machine Learning. Healthcare Analytics, 3, 100198. https://doi.org/10.1016/j.healthc.2023.100198

Ramandanis, D., and Xinogalos, S. (2023). Designing a Chatbot for Contemporary Education: A Systematic Literature Review. Information, 14, 503. https://doi.org/10.3390/info14090503

Rebelo, N., Sanders, L., Li, K., and Chow, J. C. L. (2022). Learning the Treatment Process in Radiotherapy Using an Artificial Intelligence–Assisted Chatbot: Development Study. JMIR Formative Research, 6, e39443. https://doi.org/10.2196/39443

Siddique, S., and Chow, J. C. L. (2020). Artificial Intelligence in Radiotherapy. Reports of Practical Oncology and Radiotherapy, 25, 656–666. https://doi.org/10.1016/j.rpor.2020.08.002

Siddique, S., and Chow, J. C. L. (2021). Machine Learning in Healthcare Communication. Encyclopedia, 1, 220–239. https://doi.org/10.3390/encyclopedia1010013

Varitimiadis, S., Kotis, K., Pittou, D., and Konstantakis, G. (2021). Graph-Based Conversational AI: Towards a Distributed and Collaborative Multi-Chatbot Approach for Museums. Applied Sciences, 11, 9160. https://doi.org/10.3390/app11249160

Xu, L., Sanders, L., Li, K., and Chow, J. C. (2021). Chatbot for Health Care and Oncology Applications Using Artificial Intelligence and Machine Learning: Systematic Review. JMIR Cancer, 7, e27850. https://doi.org/10.2196/27850