The DOI matches its Crossref record; 25 of the 25 references it cites were found in a registry or an index; 9 of the 17 sources its claims cite were checked for support.
Scaffolding Creativity: Integrating Generative AI Tools and Real-World Experiences in Business Education
Open accessAbstract
This exploratory study investigates the intersection of Generative AI tools and experiential learning in business education. Through a case study of an innovative undergraduate course, we examine how students interact with and adapt to various AI modalities-from text-based tools to image generation-alongside real-world experiences. Our findings reveal how this integrated approach enables novice users to overcome creative barriers, accelerates skill acquisition, and creates a dynamic interplay between AI-generated insights and real-world validation. We identify critical interaction challenges, including prompt engineering patterns and the need for more intuitive AI interfaces in educational contexts. These insights inform the design of future AI tools for creative learning and contribute to broader HCI discussions about human-AI collaboration in educational settings.
Overview
CoChat Summary
TL;DR
This exploratory case study examines how integrating generative AI tools with real-world experiences in an undergraduate business branding course affected students’ creative processes and learning outcomes. It finds that AI-assisted preparation accelerated domain-specific knowledge acquisition and supported visual concept development, while also revealing prompt-engineering challenges, AI knowledge limitations, and a shift toward individual work.
Key findings
- Students used ChatGPT-3.5 to research unfamiliar production and raw-material topics; prompt histories from 3 students showed non-linear learning trajectories shaped by prior domain knowledge, and all 8 students’ presentations showed rapid project-specific knowledge acquisition completed within 4–8 prompts.
- The Yiwu market field trip let students validate AI-generated insights: AI preparation helped them identify manufacturers on 1688.com and engage vendors, sometimes presenting as experienced buyers, but also exposed AI limitations—e.g., S1 found 100% hemp production was far less common than AI had implied, and many materials and production techniques encountered were not mentioned in AI-assisted research.
- Multimodal AI after Week 9 functioned as a cognitive scaffold for visual transformation; Midjourney interaction records from 4 students showed S2: 90 interactions, 14 concepts, 45 iterations, 2.6 effective interactions per concept; S3: 31, 10, 10, 1.0; S4: 40, 12, 19, 1.6; S5: 12, 4, 6, 1.5.
- Despite encouragement to work in pairs after the brand proposal stage, six of eight students insisted on working individually; the paper suggests AI may reduce perceived need for collaboration and raises concerns about developing interpersonal skills.
Methods
- Exploratory case study of an undergraduate course, “Reinventing the Brand,” with 8 students from six nationalities; all reported no prior formal AI literacy training, with varying ChatGPT-3.5 engagement experience.
- The course phased in ChatGPT-3.5, real-world experiences including a local mall visit and a two-day Yiwu Small Merchandise Market field trip, and a Week 9 multimodal GenAI training session covering tools such as Midjourney, Leonardo.AI, Pika, Suno.ai, and ElevenLabs.
- Data included MP4 outputs (N=8), ChatGPT prompt histories (N=3), Yiwu observation/reflection (N=8), branding proposal (N=7), Midjourney documentation (N=4), and final brand portfolio (N=8); analysis relied on instructor observations, student reflections, and qualitative comparisons rather than formal coding.
Limitations
- The paper states that formal pre/post assessments or control groups were not feasible; the sample was small; no structured qualitative analysis such as thematic analysis or inductive coding was used; prompt histories were available for only 3 students and Midjourney documentation for 4.
- The paper also states that it did not formally measure whether AI enhanced creative capacity or merely improved execution efficiency, classroom observations were not systematically logged, and comparison with a prior non-AI course was based on instructor experience rather than controlled experimental comparison.
- Evident from the text: findings are limited to a single course and small international cohort, rely heavily on instructor interpretation and self-selected documentation, and lack longitudinal or cross-disciplinary assessment.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
6 Valid · 1 Not valid · 6 Not checked · 70 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“This observation demonstrates the potential of AI as a tool for just-in-time learning in project-based contexts [8, 18], where students can quickly acquire specific expertise needed for their individual projects within compressed timelines.”
What the sources say
- Not checked — no full text of the sourcecites [8] [8] Minkyoung Kim and Lauren Adlof. 2024. Adapting to the future: ChatGPT as a means for supporting constructivist learning environments. TechTrends 68, 1 (2024), 37-46. - source
- Partly supportedcites [18] [18] Kyoungwon Seo, Joice Tang, Ido Roll, Sidney Fels, and Dongwook Yoon. 2021. The impact of artificial intelligence on learner-instructor interaction in online learning. International Journal of Educational Technology in Higher Education 18 (2021), 1-23. - sourceExperimental — accuracy not yet measured
The Findings > Support section describes students’ expectations that AI could provide just-in-time, personalized help with studying and group projects. It does not say students could quickly acquire project-specific expertise within compressed timelines; those details go beyond the paper’s findings. (LLM, conf=0.85)
boostedstates a possible causen = 3- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“The rapid adaptation to visual communication tools suggests that multimodal AI is democratizing design capabilities in non-design disciplines[1, 23].”
What the sources say
- Not checked — no full text of the sourcecites [1] [1] Adebowale Jeremy Adetayo. 2024. Reimagining learning through AI art: The promise of DALL-E and MidJourney for education and libraries. Library Hi Tech News (ahead-of-print) (2024). - source
- Supportedcites [23] [23] Hu Yin, Zipeng Zhang, and Yuanyuan Liu. 2023. The exploration of integrating the Midjourney artificial intelligence generated content tool into design systems to direct designers towards future-oriented innovation. Systems 11, 12 (2023), 566. - source
The Discussion says AIGC lowers the barrier to design collaboration for non-design professionals. It also describes students from diverse professional backgrounds using Midjourney to express design concepts rapidly and visually, supporting the paraphrase about broader access to design capabilities. (LLM, conf=0.91)
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Novice users also face significant hurdles in prompt engineering [24, 26].”
What the sources say
- Supportedcites [24] [24] J. Diego Zamfirescu-Pereira, Richmond Y. Wong, Björn Hartmann, and Qian Yang. 2023. Why Johnny can't prompt: how non-AI experts try (and fail) to design LLM prompts. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems . 1-21. - source
Section 4.2 describes substantial difficulties non-experts encountered while iterating on prompts, including over-generalizing from limited experience. The “Choosing the Right Instructions” subsection says finding effective instructions was a struggle for nearly every participant. (LLM, conf=0.94)
- Partly supportedcites [26] [26] Josef Šedlbauer, Jan Činčera, Martin Slavík, and Adéla Hartlová. 2024. Students' reflections on their experience with ChatGPT. Journal of Computer Assisted Learning 40, 4 (2024), 1526-1534. , , sourceExperimental — accuracy not yet measured
Section 3.1 reports that most students had not used generative AI before and that some struggled to formulate useful prompts. But the Discussion (§4) describes adoption as swift and relatively easy, with only some students encountering this problem. The claim overstates both how widespread and how significant the hurdle was. (LLM, conf=0.86)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“This collaborative process aligns with the vision that AI tools act not just as implementers of human ideas, but as active collaborators in the creative process [7, 23].”
What the sources say
- Supportedcites [7] [7] Jinhee Kim, Hyunkyung Lee, and Young Hoan Cho. 2022. Learning design to support student-AI collaboration: Perspectives of leading teachers for AI in education. Education and Information Technologies 27, 5 (2022), 6069-6104. - source
Section 2.1, “Student-AI collaboration” (page not provided), explicitly frames AI as a collaborative learning partner rather than a mere tool. Section 4.2.3, “Creative tasks” (page not provided), identifies writing, drawing, and music composition as activities for student-AI collaboration. Together, these support the cited vision of AI as an active collaborator in creative work. (LLM, conf=0.88)
- Supportedcites [23] [23] Hu Yin, Zipeng Zhang, and Yuanyuan Liu. 2023. The exploration of integrating the Midjourney artificial intelligence generated content tool into design systems to direct designers towards future-oriented innovation. Systems 11, 12 (2023), 566. - source
The Discussion (page not provided) describes AIGC as a co-creative platform and says it can contribute judgment, feedback, and optimization suggestions to design collaboration. That supports the paraphrase that AI does more than implement designers’ ideas, though the paper also stresses preserving the originality of human concepts. (LLM, conf=0.88)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The diverse learning trajectories observed in these cases illustrate how AI tools can facilitate non-linear, adaptive learning processes by enabling students to dynamically integrate new information with prior knowledge and external resources [15, 16].”
What the sources say
- Supportedcites [15] [15] Tareq Rasul, Sumesh Nair, Diane Kalendra, Mulyadi Robin, Fernando de Oliveira Santini, Wagner Junior Ladeira, Mingwei Sun, Ingrid Day, Raouf Ahmad Rather, and Liz Heathcote. 2023. The role of ChatGPT in higher education: Benefits, challenges, and future research directions. Journal of Applied Learning and Teaching 6, 1 (2023), 41-56. - source
The Adaptive learning section describes ChatGPT-supported learning as adaptive and scaffolded, connecting new information with prior knowledge while providing on-demand educational resources. The cited paper supports the proposed learning mechanism; the 'diverse learning trajectories observed in these cases' are the citing authors’ own observation, not a finding attributed to this source. Page numbers were not provided in our copy. (LLM, conf=0.87)
- Partly supportedcites [16] [16] Jürgen Rudolph, Samson Tan, and Shannon Tan. 2023. ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching 6, 1 (2023), 342-363. - sourceExperimental — accuracy not yet measured
The Student-facing AI applications section (page unavailable) says AI can scaffold learning in adaptive, personalised ways and provide personalised tutoring. It does not describe non-linear learning processes or students dynamically integrating new information with prior knowledge and external resources. The trajectories observed in the citing paper's cases are the citing authors' own observation, not a finding attributed to this source. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
42 more claims
Create a free account to see all 83 claims and every source verdict.
- Probably not a substantial claim. 24
- It may state two separate claims as one. 20
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“| Concepts | 14 | 10 | 12 | 4 | | Iterations | 45 | 10 | 19 | 6 | | Effective Interactions per Concept | 2.6 | 1.0 | 1.6 | 1.5 |”
Reported 2.6 · recomputed 3.214286
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Non-Linear Learning Trajectories and Knowledge Integration · p. 4
“she immediately followed up with a question [S1,P4] about the differences between cotton and hemp jeans”
§ A Student Prompt Histories and Observed Behaviors · p. 8
“| P2 | What differences are there in the production of cotton jeans and hemp jeans from a sustainability perspective? | Narrows down; shows curiosity |”
Citations
Cites
The standard definition of creativity
Runco, M. A., & Jaeger, G. J. (2012) · Creativity Research Journal · doi:10.1080/10400419.2012.650092 · Cited by 3,895 · cited 1×
“…the development of visual and creative skills Creativity, broadly defined as the ability to generate ideas that are both novel and effective [17], plays a central role in this study, as AI tools have the potential to redefine how students engage with idea generation, problem-solving, and…”
Optimal experience in work and leisure.
Csikszentmihalyi, M., & LeFevre, J. (1989) · Journal of Personality and Social Psychology · doi:10.1037//0022-3514.56.5.815 · Cited by 1,943
ChatGPT: Bullshit spewer or the end of traditional assessments in higher education?
(2023) · Journal of Applied Learning & Teaching · doi:10.37074/jalt.2023.6.1.9 · Cited by 1,806 · cited 2×
“The diverse learning trajectories observed in these cases illustrate how AI tools can facilitate non-linear, adaptive learning processes by enabling students to dynamically integrate new information with prior knowledge and external resources [15, 16].”
Generative AI and ChatGPT: applications, challenges, and AI-human collaboration
Fui-Hoon Nah, F., Zheng, R., Cai, J., Siau, K., & Chen, L. (2023) · Journal of Information Technology Case and Application Research · doi:10.1080/15228053.2023.2233814 · Cited by 1,363 · cited 1×
“This continuous recalibration between AI outputs and human experiences reflects ongoing discussions on Human-AI collaboration in learning contexts [14].”
Why Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts
Zamfirescu-Pereira, J. D., Wong, R. Y., Hartmann, B., & Yang, Q. (2023) · Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3581388 · Cited by 1,274
The impact of artificial intelligence on learner–instructor interaction in online learning
Seo, K., Tang, J., Roll, I., Fels, S., & Yoon, D. (2021) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-021-00292-9 · Cited by 788 · cited 2×
“This observation demonstrates the potential of AI as a tool for just-in-time learning in project-based contexts [8, 18], where students can quickly acquire specific expertise needed for their individual projects within compressed timelines.”
Kim, J., Lee, H., & Cho, Y. H. (2022) · Education and Information Technologies · doi:10.1007/s10639-021-10831-6 · Cited by 669 · cited 1×
“This collaborative process aligns with the vision that AI tools act not just as implementers of human ideas, but as active collaborators in the creative process [7, 23].”
Exploring the potential of using an AI language model for automated essay scoring
Mizumoto, A., & Eguchi, M. (2023) · Research Methods in Applied Linguistics · doi:10.1016/j.rmal.2023.100050 · Cited by 647
The rise of ChatGPT : Exploring its potential in medical education
Lee, H. (2024) · Anatomical Sciences Education · doi:10.1002/ase.2270 · Cited by 596 · cited 1×
“Future research should investigate how students develop AI literacy skills, including their ability to assess AI reliability, recognize biases, and refine AI-generated content critically [8, 9].”
The role of ChatGPT in higher education: Benefits, challenges, and future research directions
(2023) · Journal of Applied Learning & Teaching · doi:10.37074/jalt.2023.6.1.29 · Cited by 561
Cited by
Melanou, C., & Beege, M. (2026) · Education Sciences · doi:10.3390/educsci16040651 · Cited by 12
The anonymity paradox: Navigating face-saving and credibility in Chinese EFL peer feedback practices
Wang, F., Wang, M., Wu, D., & Qian, W. (2026) · Learning, Culture and Social Interaction · doi:10.1016/j.lcsi.2026.101014 · Cited by 3
Analysing Social Media as a Digital Tool for Inclusive Higher Education in India
Tripathi, S. N., Basera, J., Chopra, A., & Bhatia, V. (2025) · IEEE DELCON - International Conference on Recent Smart Technologies in Engineering for Sustainable Development (DELCON) · doi:10.1109/delcon68055.2025.11400175 · Cited by 3
“[45] This gap hampers societal movement and economic power, further limiting the complete potential of digital education[46]”
Social stereotypes in AI text-to-image generation
Barve, S., Mao, A., Shi, J. M., Juneja, P., & Saha, K. (2026) · AI and Ethics · doi:10.1007/s43681-026-01146-8 · Cited by 18
Zhou, Y., Babu, P. M., Pandey, U., Magana, A. J., & Li, T. (2026) · ACM International Conference on Intelligent User Interfaces (IUI) · doi:10.1145/3742413.3789139 · Cited by 8
Metadata
- Authors
- Wang, N. C.
- Affiliations
- New York University Shanghai (CN)
- Corresponding author
- Wang, Nicole C.
- ORCID
- Wang, Nicole C.
- Venue
- Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems
- Publisher
- ACM
- Published
- 2025-04-25
- Pages
- 1-9
- OpenAlex
- W4409720420
- Semantic Scholar
- 6cd35ae3d199dc3a7d69ae1fde4dbc806921cd55
- Cited by
- 18
- References
- 22
- Citations by year
- 2026: 14 · 2025: 4
- FWCI
- 20.58
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Topics
- Social Sciences › Education › Artificial Intelligence in Education
- Social Sciences › Education › Management and Marketing Education
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Creativity in Education and Neuroscience
- Keywords
- generative AI in education, business education, experiential learning, creativity, undergraduate course design, prompt engineering, AI interfaces, human-AI collaboration, image generation, skill acquisition, real-world validation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Scaffolding Creativity: Integrating Generative AI Tools and Real-World Experiences in Business Education. https://app.cochat.ai/papers/10.1145/3706599.3720283
- The paper
- Wang, N. C. (2025). Scaffolding Creativity: Integrating Generative AI Tools and Real-World Experiences in Business Education. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3706599.3720283
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.