Exploring Generative Models with Middle School Students
Open accessThe DOI matches its Crossref record; 37 of the 37 references it cites were found in a registry or an index.
Abstract
Applications of generative models such as Generative Adversarial Networks (GANs) have made their way to social media platforms that children frequently interact with. While GANs are associated with ethical implications pertaining to children, such as the generation of Deepfakes, there are negligible efforts to educate middle school children about generative AI. In this work, we present a generative models learning trajectory (LT), educational materials, and interactive activities for young learners with a focus on GANs, creation and application of machine-generated media, and its ethical implications. The activities were deployed in four online workshops with 72 students (grades 5-9). We found that these materials enabled children to gain an understanding of what generative models are, their technical components and potential applications, and benefits and harms, while reflecting on their ethical implications. Learning from our findings, we propose an improved learning trajectory for complex socio-technical systems.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures










Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“This objective develops students' understanding of how AI works, their ability to assess truthfulness, detect bias, and recognize the limits of creativity.”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [134]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Parents and teachers likewise raise questions about age appropriateness, ethical risk, and the costs of integration [103,116,121,123,134].”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [134]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Our findings confirm that GenAI enhances motivation and self-efficacy, consistent with Ali et al. [134] previous work.”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [134]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In our review, a few papers [2,6,23,30] explicitly raised concerns about the opaque nature of AI tools, with children and adults questioning where the AI-generated information comes from, why certain outputs were created, or how the system made decisions.”
From Child-AI Co-Creation: A Review of the Current Research Landscape and a Proposal for Six Design Considerations (2025) · cites this paper as [2]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, due to the fast-paced, rapidly changing nature of the technology, the usability of generative AI tools and their applicability in educational settings has been under-explored [2].”
From Design implications of generative AI systems for visual storytelling for young learners (2023) · cites this paper as [2]
Experimental — accuracy not yet measured
3 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
GAN(Generative Adversarial Nets)
(2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 19,270
(2001) · Cited by 13,930
Cognitive Load During Problem Solving: Effects on Learning
Sweller, J. (1988) · Cognitive Science · doi:10.1016/0364-0213(88)90023-7 · Cited by 11,048 · cited 1×
“Eforts were made to decrease cognitive load [50] and reduce the space of possibilities during game play.”
BackgroundWhat Video Games Have to Teach Us about Learning and Literacy
(2004) · Education + Training · doi:10.1108/et.2004.00446dae.002 · Cited by 6,674
Analyzing and Improving the Image Quality of StyleGAN
Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., & Aila, T. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00813 · Cited by 6,332
Analyzing and Improving the Image Quality of StyleGAN
Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., & Aila, T. (2019) · arXiv · doi:10.48550/arxiv.1912.04958 · Cited by 6,332 · cited 1×
“• This Person Does Not Exist: This tool utilizes StyleGan2 [29] and has been trained on human faces to generate fake human faces using GANs [53].”
MethodWhat video games have to teach us about learning and literacy
Gee, J. P. (2003) · Computers in Entertainment · doi:10.1145/950566.950595 · Cited by 5,735 · cited 1×
“We made use of game-based learning approaches to make abstract concepts concrete since games have shown to enhance content engagement [19, 20, 31].”
MethodWhat is AI Literacy? Competencies and Design Considerations
Long, D., & Magerko, B. (2020) · Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3313831.3376727 · Cited by 3,241 · cited 1×
“Long and Magerko (2020) described AI literacy as “AI literacy as a set of competencies that enables individuals to critically evaluate AI technologies; communicate and collaborate efectively with AI; and use AI as a tool online, at home, and in the workplace" [36].”
BackgroundA Neural Algorithm of Artistic Style
Gatys, L., Ecker, A., & Bethge, M. (2016) · Journal of Vision · doi:10.1167/16.12.326 · Cited by 2,612
Winner, L. (2017) · Computer Ethics · doi:10.4324/9781315259697-21 · Cited by 2,317 · cited 1×
“However, in the case of knowledge about socio-technical systems (systems that constitute interactions between society’s complex infrastructures and human behaviour technical components), applications of the technology, and ethical implications are interconnected and inform each other [58].”
BackgroundTutorial on Variational Autoencoders
Doersch, C. (2016) · arXiv · doi:10.48550/arxiv.1606.05908 · Cited by 1,470 · cited 1×
“There are several diferent kinds of generative models such as Generative Adversarial Networks (GANs) [21], Variational Autoencoders (VAEs) [13], and Autoregressive models such as PixelRNN [41].”
Pixel Recurrent Neural Networks
van den Oord, A., Kalchbrenner, N., & Kavukcuoglu, K. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1601.06759 · Cited by 1,437 · cited 1×
“There are several diferent kinds of generative models such as Generative Adversarial Networks (GANs) [21], Variational Autoencoders (VAEs) [13], and Autoregressive models such as PixelRNN [41].”
The common sense census: media use by tweens and teens
Rideout, V. (2015) · Cited by 1,098
Envisioning AI for K-12: What Should Every Child Know about AI?
Touretzky, D., Gardner-McCune, C., Martin, F., & Seehorn, D. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33019795 · Cited by 918 · cited 1×
“Unsurprisingly, several new AI curricula and frameworks have been developed for k12 learners [56].”
Background(2018) · Cited by 900
Grounded Theory and Theoretical Coding
Thornberg, R., & Charmaz, K. (2014) · The SAGE Handbook of Qualitative Data Analysis · doi:10.4135/9781446282243.n11 · Cited by 741 · cited 1×
“We used the grounded theory of coding, where a coding scheme were formed from the data collected by two coders and validated using an inter-coder reliability (IR = 0.84) [54].”
MethodInfluentialKlopfer, E. D., Osterweil, S., & Salen, K. (2009) · HAL · Cited by 557
GANs for medical image analysis
Kazeminia, S., Baur, C., Kuijper, A., van Ginneken, B., Navab, N., Albarqouni, S., & Mukhopadhyay, A. (2020) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2020.101938 · Cited by 534
A Neural Representation of Sketch Drawings
Ha, D., & Eck, D. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.03477 · Cited by 496 · cited 1×
“• Sketch RNN: Built by Google Creative Lab, Sketch RNN is an interactive web experiment that lets you draw together with a recurrent neural network model [25].”
BackgroundDeveloping Middle School Students' AI Literacy
Lee, I., Ali, S., Zhang, H., DiPaola, D., & Breazeal, C. (2021) · Proceedings of the 52nd ACM Technical Symposium on Computer Science Education · doi:10.1145/3408877.3432513 · Cited by 373 · cited 1×
“In our workshop, students completed a pre-test questionnaire before the workshop, and a post-test questionnaire again after the workshop as part of a larger CI developed around AI literacy [34].”
MethodSelf-citationAssessing social-emotional development in children from a longitudinal perspective
Denham, S. A., Wyatt, T. M., Bassett, H. H., Echeverria, D., & Knox, S. S. (2009) · Journal of Epidemiology and Community Health · doi:10.1136/jech.2007.070797 · Cited by 315 · cited 1×
“In terms of thinking and learning, middle school students are able to think more abstractly than younger students [2, 12] and they can incorporate new knowledge into existing schemas [27, 28].”
BackgroundTeachable Machine: Approachable Web-Based Tool for Exploring Machine Learning Classification
Carney, M., Webster, B., Alvarado, I., Phillips, K., Howell, N., Griffith, J., Jongejan, J., Pitaru, A., & Chen, A. (2020) · Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3334480.3382839 · Cited by 313 · cited 1×
“Topics of learning included basics of machine learning [39], audio and image classifcation through creative programming tools [10, 26, 52], data science [49], clustering algorithms [57], and AI ethics [4, 42].”
Schawinski, K., Zhang, C., Zhang, H., Fowler, L., & Santhanam, G. K. (2017) · Monthly Notices of the Royal Astronomical Society: Letters · doi:10.1093/mnrasl/slx008 · Cited by 246 · cited 1×
“While largely used for experimental generation of art and entertainment [18], GANs have applications in science such as improving astronomical images [47] or detecting glaucomatous images that help with early diagnosis of blindness [8].”
MethodTeaching Machine Learning in School: A Systematic Mapping of the State of the Art
MARQUES, L. S., GRESSE VON WANGENHEIM, C., & HAUCK, J. C. R. (2020) · Informatics in Education · doi:10.15388/infedu.2020.14 · Cited by 170 · cited 1×
“Topics of learning included basics of machine learning [39], audio and image classifcation through creative programming tools [10, 26, 52], data science [49], clustering algorithms [57], and AI ethics [4, 42].”
Progress on concept inventory assessment tools
Evans, D. L., Gray, G. L., Krause, S., Martin, J., Midkiff, C., Notaroš, B. M., Pavelich, M. J., Rancour, D., Reed, T., Steif, P. S., et al. (2004) · IEEE Frontiers in Education Conference (FIE) · doi:10.1109/fie.2003.1263392 · Cited by 165
Learning trajectories: a framework for connecting standards with curriculum
Confrey, J., Maloney, A., & Corley, A. K. (2014) · ZDM · doi:10.1007/s11858-014-0598-7 · Cited by 129 · cited 1×
“LTs have been extensively used for building curricula in mathematics [11], and more recently in Computer Science [17].”
MethodDeep Learning for Deepfakes Creation and Detection: A Survey
Nguyen, T. T., Nguyen, M. C., Nguyen, D. T., Nguyen, D. T., & Nahavandi, S. (2020) · Figshare · doi:10.6084/m9.figshare.12731039.v1 · Cited by 121
Deep learning for deepfakes creation and detection: a survey
Nguyen, T. T., Nguyen, Q. V. H., Nguyen, D. T., Nguyen, D. T., Huynh-The, T., Nahavandi, S., Nguyen, T. T., Pham, Q.-V., & Nguyen, C. M. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4030341 · Cited by 112
Games: Purpose and Potential in Education
Miller, C. T. (2008) · Springer eBooks · doi:10.1007/978-0-387-09775-6 · Cited by 88
SmileyCluster: supporting accessible machine learning in K-12 scientific discovery
Wan, X., Zhou, X., Ye, Z., Mortensen, C. K., & Bai, Z. (2020) · Proceedings of the Interaction Design and Children Conference · doi:10.1145/3392063.3394440 · Cited by 77 · cited 1×
“Topics of learning included basics of machine learning [39], audio and image classifcation through creative programming tools [10, 26, 52], data science [49], clustering algorithms [57], and AI ethics [4, 42].”
Battista, M. T. (2011) · The Mathematics Enthusiast · doi:10.54870/1551-3440.1228 · Cited by 75 · cited 2×
“…presented in the paper were developed after considering the theoretical background of LTs [48], Battista (2011) defnes a Learning Trajectory (LT) as “a detailed description of the sequence of thoughts, ways of reasoning, and strategies that a student employs while involved in learning a topic" [7].”
BackgroundChild-Friendly Programming Interfaces to AI Cloud Services
Kahn, K., & Winters, N. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-66610-5_64 · Cited by 74 · cited 1×
“Topics of learning included basics of machine learning [39], audio and image classifcation through creative programming tools [10, 26, 52], data science [49], clustering algorithms [57], and AI ethics [4, 42].”
Generative adversarial network and texture features applied to automatic glaucoma detection
Bisneto, T. R. V., Filho, A. O. D. C., & Magalhães, D. M. V. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106165 · Cited by 66 · cited 1×
“While largely used for experimental generation of art and entertainment [18], GANs have applications in science such as improving astronomical images [47] or detecting glaucomatous images that help with early diagnosis of blindness [8].”
MethodIntroducing Data Science to School Kids
Srikant, S., & Aggarwal, V. (2017) · Proceedings of the 2017 ACM SIGCSE Technical Symposium on Computer Science Education · doi:10.1145/3017680.3017717 · Cited by 62 · cited 1×
“Topics of learning included basics of machine learning [39], audio and image classifcation through creative programming tools [10, 26, 52], data science [49], clustering algorithms [57], and AI ethics [4, 42].”
Evaluation of a learning trajectory for length in the early years
Sarama, J. H., Clements, D. H., Barrett, J. E., Dine, D. W. V., & McDonel, J. S. (2011) · ZDM · doi:10.1007/s11858-011-0326-5 · Cited by 54 · cited 2×
“Hypothetical Learning Trajectories (HLTs) were useful in articulating how learning goals, cognitive tasks; and instructional tasks ft together and formed a theory of action in design research [46].”
MethodExploring Quantum Reversibility with Young Learners
Franklin, D. M., Palmer, J., Jang, W., Lehman, E., Marckwordt, J., Landsberg, R. H., Muller, A., & Harlow, D. B. (2020) · ACM Conference on International Computing Education Research (ICER) · doi:10.1145/3372782.3406255 · Cited by 21 · cited 2×
“Learning trajectories have proven helpful when taking a constructivist approach to curriculum development wherein students’ existing knowledge supplements learning new concepts [17].”
BackgroundMethodLearning from Multi-domain Artistic Images for Arbitrary Style TransferPreprint
Xu, Z., Wilber, M. J., Chen, F., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.09987 · Cited by 8
Research works printed but not linked: 8
- 3. Safnah Ali, Hae Won Park, and Cynthia Breazeal. 2020. A social robot's infuence on children's fgural creativity during gameplay. International Journal of Child- Computer Interaction (2020), 100234.
- 9. Benjamin Samuel Bloom. 1956. Taxonomy of educational objectives: The classif- cation of educational goals. Cognitive domain (1956).
- 20. Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In Advances in neural information processing systems. 2672-2680.
- 37. Yotam Mann. 2016. AI duet. Experiments with Google. See, https://experiments. withgoogle. com/ai/ai-duet (2016).
- 39. Thanh Thi Nguyen, Cuong M Nguyen, Dung Tien Nguyen, Duc Thanh Nguyen, and Saeid Nahavandi. 2019. Deep learning for deepfakes creation and detection. arXiv preprint arXiv:1909.11573 (2019). CHI '21, May 8-13, 2021, Yokohama, Japan · arXiv
- 41. B.H. Payne. 2020. Can my algorithm be my opinion?: An AI + Ethics Curriculum for Middle School Students. Master's thesis. Massachusetts Institute of Technology, Media Lab, Cambridge, MA, USA.
- 47. Martin A Simon. 1995. Reconstructing mathematics pedagogy from a construc- tivist perspective. Journal for research in mathematics education (1995), 114-145.
- 51. Danny Tang et al. 2019. Empowering novices to understand and use machine learning with personalized image classifcation models, intuitive analysis tools, and MIT App Inventor. Ph.D. Dissertation. Massachusetts Institute of Technology.
Cited by
Bahroun, Z., Anane, C., Ahmed, V., & Zacca, A. (2023) · Sustainability · doi:10.3390/su151712983 · Cited by 784
“This section, therefore, covers a review of a wide range of applications and considerations regarding the integration of GAI in education, as discussed in Table 8 below: ) [117].”
BackgroundAI literacy in K-12: a systematic literature review
Casal-Otero, L., Catala, A., Fernández-Morante, C., Taboada, M., Cebreiro, B., & Barro, S. (2023) · International Journal of STEM Education · doi:10.1186/s40594-023-00418-7 · Cited by 724
“…how a robot equipped with AI components can be used in society (Eguchi & Okada, 2018), program conversational agents (Van Brummelen et al., 2021b), or learn to recognize credible but fake media products (video, photos), which have been generated using AI-based techniques (2021b; Ali et al., 2021a).”
BackgroundZhang, H., Lee, I., Ali, S., DiPaola, D., Cheng, Y., & Breazeal, C. (2023) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-022-00293-3 · Cited by 396
“The three core domains were established by the project leadership team and informed by our team’s previous research on promoting AI education among young learners (e.g., Ali et al., 2019, 2021; DiPaola et al., 2020).”
MethodSelf-citationForero-Corba, W., & Bennásar, F. N. (2023) · RIED Revista Iberoamericana de Educación a Distancia · doi:10.5944/ried.27.1.37491 · Cited by 371
Scope of machine learning in materials research—A review
Mobarak, M. H., Mimona, M. A., Islam, M. A., Hossain, N., Zohura, F. T., Imtiaz, I., & Rimon, M. I. H. (2023) · Applied Surface Science Advances · doi:10.1016/j.apsadv.2023.100523 · Cited by 230
Metadata
- Authors
- Ali, S., DiPaola, D., Lee, I., Hong, J., & Breazeal, C.
- Affiliations
- MIT Media Lab (US)
- Massachusetts Institute of Technology (US)
- Venue
- Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems
- Publisher
- ACM
- Published
- 2021-05-06
- Pages
- 1-13
- OpenAlex
- W3161501253
- Semantic Scholar
- f12664a05fab765fdc4e159689dae76bca15e728
- Cited by
- 82
- References
- 30
- Citations by year
- 2026: 11 · 2025: 24 · 2024: 22 · 2023: 18 · 2022: 5 · 2021: 2
- FWCI
- 9.74
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- CHI Conference on Human Factors in Computing Systems (publishedVersion)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
- Funders
- National Science Foundation (2022502)
- Amazon Future Engineer
- Topics
- Social Sciences › Education › Artificial Intelligence in Education
- Physical Sciences › Computer Science › Computer Science Applications › Teaching and Learning Programming
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Keywords
- generative models, generative adversarial networks, junior high school students, deepfakes, AI ethics, learning trajectories, teaching materials, online workshops, socio-technical systems, social media use by children
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.
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
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.