Daniel Zviel Girshin

ILAIS 2025, 19th ILAIS Conference, July 7, 2025, Ben-Gurion University

 

Multi-agent GenAI-Human Dialogue Ecosystem as Information Source

Daniel Zviel Girshin, AI and Robotics Disruptor Lab, Ruppin

Abstract

GenAI is structured and exact but not too original. People are inventive and unique but less structured. Information from these as a stand-alone source is lacking. GenAI generating depends heavily on human creativity in the prompt. Dialogue, chain of prompts-answers is a better source. An ecosystem of many competing dialogue chains is richer. Evolution of such ecosystem of multitude of human and artificial agents creates, through competition in real life, a new revolutionary source of information. It also creates a lot of information about users, processes, and other metadata that is used for knowledgebase. An additional layer of human monitoring, management, experiencing, experimenting and guiding creates a better database and knowledgebase as explicit and implicit side-effects. The next layer is introducing an element of annealing, extreme creativity, non-conformism, and brainstorming - a very productive process, that will be called by us - disruption. In the next layer are mechanisms to transform it into a practical, efficient, stable, learning and productive creative ecosystem. We exemplify this by presenting an ecosystem realizing this paradigm, that we have built in our Engineering Disruption Lab for the last few years for engineering students educational needs – PAPERT.

Keywords: GenAI, multi-agent system, engineering disruption, education.

Introduction

Generative Artificial intelligence (AI), by definition can (and does) create a wealth of information, and by some measure even knowledge. This could become a useful, more structured, but less original information production, while human cognition tends to be more inventive but less organized. Artificial intelligence, even the generative models, exhibit a unique characteristic: highly structured but not really original outputs. A simple but pithy example was given by Narendra Damodardas Modi, Prime Minister of India, while co-chairing an AI Action Summit in Paris in February 2025. He argued from his own experience that when prompting an AI to generate images depicting handwriting, even when explicitly indicating that the writer is left-handed, the user will get an image of right-handed writer.

On the other hand, humans as sources of information are more important, and they are inherently inventive, producing less structured and more diverse outputs. But people as direct, unfiltered and unprocessed sources of information are extremely unreliable and hard to transform into a good useful database. It is enough to see what is said in the social networks, to understand how big a wealth of material is there, but on the other hand, how unreliable, and even scandalous, a big chunk of it is.

This dichotomy poses a challenge when attempting to leverage AI for creative information and knowledge-generating tasks. This paper describes a system developed in our engineering disruptor lab that explores the potential of combining AI's structured knowledge with human creativity to form dynamic ecosystems of knowledge generation. Drawing on theories from information systems, cognitive science, and creativity studies, we have built a model and platform for an innovative framework for multi-agent AI-human dialogue systems. We also address the challenges of maintaining stability and productivity within such ecosystems by incorporating special mechanisms that enhance creative disruption without leading to inefficiency.

Theory

The evolutionary multi-agent ecosystem is a dynamic interaction between AI and human agents that can be conceptualized as an evolutionary process (Şenyüz et al.,2025). Drawing from evolutionary computing theories (Mitchell, 1998), we can consider each dialogue chain as a potential solution that adapts and mutates through interaction. By introducing competitive and cooperative dynamics, these chains evolve, producing increasingly refined and novel knowledge structures. However, the unregulated proliferation of ideas can also result in inefficiency and instability. Therefore, implementing a layer of human monitoring and management becomes essential. Human oversight can guide the dialogue chains, refining, consolidating, and discarding less promising pathways, thereby maintaining stability while preserving creativity.

Disruption is one crucial element of such a creative ecosystem — a phase characterized by extreme creativity, non-conformism, and brainstorming (Lile, 2024). We can call this aspect 'annealing’, drawing from the metaphor of simulated annealing in optimization algorithms (Kirkpatrick et al., 1983). During the disruption phase, the system temporarily becomes less structured, allowing for unconventional ideas to emerge. To ensure the ecosystem remains productive, mechanisms must be in place to transition from disruption back to structured, goal-oriented development. These mechanisms might include filtering redundant ideas, clustering related concepts, and evaluating innovative propositions for feasibility.

Method

Balancing creativity and stability, managing the inherent tension between creative freedom and system stability requires a multi-layered approach. Firstly, introducing adaptive algorithms that monitor the quality and coherence of dialogue chains can prevent chaotic proliferation. Secondly, human curation and iterative refinement can help anchor creative ideas within a pragmatic framework. Finally, incorporating feedback loops that assess the relevance and originality of the generated content ensures that the ecosystem remains aligned with its knowledge-generation goals. Crowdsourcing, especially structured crowdsourcing, is a method for multitude of human agents’ involvement. The comprehensive database records every event, including every prompt, GenAI answer, teacher communication, peer communication, quizzes, formative and summative assessment.

Application

PAPERT (Pro-Active PERsonal Tutor) is already in use by our students.

Use-Case Examples

Passive

Student is given passive (non-interactive, asynchronous) material by the teacher (PowerPoint slides, text, online resources in various formats), and a sequence of prompts to ask, read, think and add to his knowledge. Some prompts could be mandatory.

Student is encouraged to:

·       elaborate on every prompt,

·       ask additional questions,

·       ask for adjustments and customization,

·       ask more and different prompts.

Active:

·       write me a code performing a given task,

·       explain the code or its parts,

·       comment the code at a given level,

·       translate into another language,

·       give examples,

·       indicate the importance.

Among its components:

·       Learning materials for programming courses

·       GUI interface for:

o   profile of the student

o   pre-processed prompts and answers

o   roadmaps

o   context choice (style, brevity, goal etc.)

o   IDE integration

o   Forum (teacher monitored) – public and private, anonymized and otherwise

o   Choice of course and subject

o   Text-to-Speech and Speech-to-Text

·       ML and Statistical processing of the DB including:

o   Clustering of students, prompts, answers, teacher communications etc.

o   Correlations

o   ANOVA

o   Using the results to improve the environment and create recommenders

·       Reports to teacher of summaries of student and class, including with graphs

·       Feedback through questionnaires, quizzes automatically generated for summative and formative assessments

Protocols of dialogue between student and tutor of questions and answers, some in code, some in the format of:

1.     Teach about the subject-matter (general background, color, description, motivation)

a.      the paradigm:

                                               i.     aims

                                             ii.     terminology

                                            iii.     ontology

                                            iv.     epistemology

                                              v.     techniques

                                            vi.     examples

                                           vii.     cases

                                         viii.     principles

                                            ix.     intra and inter connections and associations

                                              x.     typical use cases

                                            xi.     resources

                                           xii.     axioms and opinions

                                         xiii.     sociology and personality

2.     open or closed options dialog

a.      student is asked about his opinions

b.     student is offering his opinion

c.      evaluation and analysis of student’s opinion

3.     search resources dialog

a.      student is proposed resources

b.     student is offering resources

c.      evaluation and analysis of student’s proposal of resource

d.     reading together

e.      dialog about content and meta-data about the resource

4.     reading and understanding (any resource, but for brevity’s sake we will call it text)

a.      student is proposing a text

b.     student is offered a text

c.      student reads the text

d.     student asks about the text

e.      student writes his version, summary, opinion

f.      evaluation and analysis of student’s opinion

5.     code

a.      writing

b.     reading

c.      evaluation and analysis of student’s code and analysis

d.     improvements

e.      explanations and comments

6.     playing, tinkering, making, analyzing lab

a.      student is asked about his opinions

b.     student is offering his opinion

c.      evaluation and analysis of student’s opinion

7.     project engineering experimentalizing (lab)

a.      algorithms

b.     program logic

c.      program overview and structure

d.     approaches (like functional, OOP)

e.      engineering principles

f.      life cycles

g.     know-how for every stage

Conclusions

The approach of creating excellent sources of information through building multi-agent ecosystems of human and artificial agents with chains of prompt dialogues with GenAI has proved to be extremely successful.

It could be implemented in every domain, including, for example:

• Healthcare: Assisting in diagnosis, summarizing medical literature, and improving patient communication.
• Education: Personalizing learning experiences and enabling intelligent tutoring systems.
• Creative Arts: Generating poetry, music, and visual art, pushing the boundaries of creativity.
• Business: Automating customer support, drafting documents, and enhancing productivity tools.

References

1.Kirkpatrick, S., Gelatt, C. D., & Vecchi, M. P. 1983. Optimization by Simulated Annealing. Science, 220(4598), 671-680.

2.Lile, S., Ansari, S. and Urmetzer, F., 2024. Rethinking disruptive innovation: unravelling theoretical controversies and charting new research frontiers. Innovation, pp.1-23.

3.Mitchell, M. 1998. An Introduction to Genetic Algorithms. MIT Press.

4.Şenyüz, B., Özgen, E. and Oğuzcan, A.U., 2025. Integrating GenAI into communication education for ‘generation prompt’: an exploration of academics' perspectives on its benefits, challenges, and future prospects in Turkiye. Encontros Bibli: revista eletrônica de biblioteconomia e ciência da informação, 30, pp.1-25.

5.Braun A. (2025). Two Perspectives on AI’s Future: Rapid Revolution vs. Normal Technology. https://www.linkedin.com/pulse/two-perspectives-ais-future-rapid-revolution-vs-normal-braun-letjf.

6.Huang, S., Lai, X., Ke, L., Li, Y., Wang, H., Zhao, X., ... & Wang, Y. (2024). AI technology panic—is AI dependence bad for mental health? A cross-lagged panel model and the mediating roles of motivations for AI use among adolescents. Psychology Research and Behavior Management, 2024 Mar 12;17:1087–1102. doi: 10.2147/PRBM.S440889. PMCID: PMC10944174  PMID: 38495087.

7.Kokotajlo Daniel , Scott Alexander, Thomas Larsen, Eli Lifland, Romeo Dean (2025). AI 2027. https://ai-2027.com. 

8. OpenAI. DALL·E: Creating Images from Text. https://openai.com/index/dall-e/, 2023. Accessed: 2025-10-28.

9. Rombach, R.; Blattmann, A.; Lorenz, D.; Esser, P.; Ommer, B. High-Resolution Image Synthesis with Latent Diffusion Models. 06

2022, pp. 10674–10685. https://doi.org/10.1109/CVPR52688.2022.01042.

10. Analytix Labs. Generative AI vs. Traditional AI: Understand Key Differences, 2024. Accessed: 2025-10-28.

11. Schneider, J. Explainable Generative AI (GenXAI): A Survey, Conceptualization, and Research Agenda. Artificial Intelligence

Review 2024, 57. https://doi.org/10.1007/s10462-024-10916-x.

12. Saarela, M.; Podgorelec, V. Recent applications of Explainable AI (XAI): A systematic literature review. Applied Sciences 2024,

14, 8884. https://doi.org/10.3390/app14198884.

13. Bennetot, A.; Donadello, I.; El Qadi El Haouari, A.; Dragoni, M.; Frossard, T.; Wagner, B.; Sarranti, A.; Tulli, S.; Trocan, M.; Chatila,

R.; et al. A Practical Tutorial on Explainable AI Techniques. ACM Comput. Surv. 2024, 57. https://doi.org/10.1145/3670685.

14. Dwivedi, R.; Dave, D.; Naik, H.; Singhal, S.; Omer, R.; Patel, P.; Qian, B.; Wen, Z.; Shah, T.; Morgan, G.; et al. Explainable AI (XAI):

Core Ideas, Techniques, and Solutions. ACM Comput. Surv. 2023, 55. https://doi.org/10.1145/3561048.

15. Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.;

Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. International Journal of

Surgery 2021, 88, n71.

16. Sengar, S.S.; Hasan, A.B.; Kumar, S.; Carroll, F. Generative artificial intelligence: a systematic review and applications. Multimedia

Tools and Applications 2024. https://doi.org/10.1007/s11042-024-20016-1.

17. Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, L.u.; Polosukhin, I. Attention is All you Need.

In Proceedings of the Advances in Neural Information Processing Systems; Guyon, I.; Luxburg, U.V.; Bengio, S.; Wallach, H.;

Fergus, R.; Vishwanathan, S.; Garnett, R., Eds. Curran Associates, Inc., 2017, Vol. 30. 18. Kingma, D.P.; Welling, M.; 2019.

18. Kingma, D.P.; Welling, M.; 2019.

19. Ray, P.P. ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations, and future scope. Internet of Things and Cyber-Physical Systems 2023, 3, 121–154. https://doi.org/10.1016/j.iotcps.2023.04.003.

20. OpenAI. Introducing ChatGPT, 2022. Accessed: 28-Oct-2025.

21. Google. Gemini FAQ, 2024. Accessed: 28-Oct-2025.

22. Google. An important next step on our AI journey, 2024. Accessed: 28-Oct-2025.

23. MidJourney. MidJourney home, 2024. Accessed: 28-Oct-2025.

24. OpenAI. DALL·E, 2024. Accessed: 28-Oct-2025.

25. Isola, P.; Zhu, J.Y.; Zhou, T.; Efros, A.A. In Proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 1125–1134.

26. Karras, T.; Laine, S.; Aila, T. A Style-Based Generator Architecture for Generative Adversarial Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 2021, 43, 4217–4228. https://doi.org/10.1109/TPAMI.2020.2970919.

27. Suno. Suno home, 2024. Accessed: 28-Oct-2025.

28. RunwayML. RunwayML home, 2024. Accessed: 28-Oct-2025.

29. Lent, M.; Fisher, W.; Mancuso, M. An Explainable Artificial Intelligence System for Small-unit Tactical Behavior. 01 2004, pp. 900–907.

30. Barredo Arrieta, A.; Díaz-Rodríguez, N.; Del Ser, J.; Bennetot, A.; Tabik, S.; Barbado, A.; Garcia, S.; Gil-Lopez, S.; Molina, D.; Benjamins, R.; et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion 2020, 58, 82–115. https://doi.org/https://doi.org/10.1016/j.inffus.2019.12.012.

31. A. Ahmed, N.; Alpkocak, A. A quantitative evaluation of explainable AI methods using the depth of decision tree. Turkish Journal of Electrical Engineering and Computer Sciences 2022, 30, 2054–2072. https://doi.org/10.55730/1300-0632.3924.

32. Miller, T. Explanation in artificial intelligence: Insights from the social sciences. Artificial Intelligence 2019, 267, 1–38. https: //doi.org/10.1016/j.artint.2018.07.007.

33. Larsson, S.; Heintz, F. Transparency in artificial intelligence. Internet Policy Review 2020, 9. https://doi.org/10.14763/2020.2.1469.

34. Braun, V.; Clarke, V. Using thematic analysis in psychology. Qualitative Research in Psychology 2006, 3, 77–101, [https://www.tandfonline.com/doi/pdf/10.1191/1478088706qp063oa]. https://doi.org/10.1191/1478088706qp063oa. 63527 of 3035. Da’u, A.; Salim, N. Recommendation system based on deep learning methods: a systematic review and new directions. Artificial Intelligence Review 2020, 53, 2709–2748. https://doi.org/10.1007/s10462-019-09744-1.

36. Kitchenham, B.; Charters, S. Guidelines for Performing Systematic Literature Reviews in Software Engineering. EBSE Technical Report EBSE-2007-01, Software Engineering Group, School of Computer Science and Mathematics, Keele University, Keele, UK, 2007.

37. Saarela, M.; Kärkkäinen, T. Can we automate expert-based journal rankings? Analysis of the Finnish publication indicator. Journal of informetrics 2020, 14, 101008. https://doi.org/10.1016/j.joi.2020.101008.

38. Bushey, J. AI-Generated Images as an Emergent Record Format. In Proceedings of the 2023 IEEE International Conference on Big Data (BigData), 2023, pp. 2020–2031.

39. Hanif, A.; Beheshti, A.; Benatallah, B.; Zhang, X.; Habiba.; Foo, E.; Shabani, N.; Shahabikargar, M. A Comprehensive Survey of Explainable Artificial Intelligence (XAI) Methods: Exploring Transparency and Interpretability. In Proceedings of the Web

Information Systems Engineering; Zhang, F.; Wang, H.; Barhamgi, M.; Chen, L.; Zhou, R., Eds. Springer Nature Singapore, 2023, pp. 915–925.

40. Zarghami, S.; Kouchaki, H.; Yang, L.; Martinez, P. Explainable Artificial Intelligence in Generative Design for Construction. 07 2024. 650

41. Longo, L.; Brcic, M.; Cabitza, F.; Choi, J.; Confalonieri, R.; Ser, J.D.; Guidotti, R.; Hayashi, Y.; Herrera, F.; Holzinger, A.; et al. Explainable Artificial Intelligence (XAI) 2.0: A manifesto of open challenges and interdisciplinary research directions. Information Fusion 2024, 106, 102301. 653

42. Mudabbiruddin, M.; Mosavi, A.; Imre, F. From Deep Learning to ChatGPT for Materials Design. In Proceedings of the 2024 IEEE 11th International Conference on Computational Cybernetics and Cyber-Medical Systems (ICCC), 2024, pp. 1–8.

43. Pan, S.; Luo, L.; Wang, Y.; Chen, C.; Wang, J.; Wu, X. Unifying Large Language Models and Knowledge Graphs: A Roadmap. IEEE Transactions on Knowledge and Data Engineering 2024, 36, 3580–3599. https://doi.org/10.1109/TKDE.2024.3352100.

44. Jain, R.; Jain, A., Generative AI in Writing Research Papers: A New Type of Algorithmic Bias and Uncertainty in Scholarly Work; 2024; pp. 656–669. 659

45. Zeiser, T.; Ehret, D.; Lutz, T.; Saar, J. Explainable AI in Manufacturing. In Proceedings of the 2024 IEEE International Conference on Engineering, Technology, and Innovation (ICE/ITMC), 2024, pp. 1–8.

46. Qu, T.; Yang, Z. Overview of Artificial Intelligence Applications in Educational Research. In Proceedings of the Proceedings of the 2024 International Symposium on Artificial Intelligence for Education, New York, NY, USA, 2024; ISAIE ’24, p. 101–108.

47. Bui, L.V. Advancing patent law with generative AI: Human-in-the-loop systems for AI-assisted drafting, prior art search, and multimodal IP protection. World Patent Information 2025, 80, 102341.

48. Ye, X.; Yigitcanlar, T.; Goodchild, M.; Huang, X.; Li, W.; Shaw, S.L.; Fu, Y.; Gong, W.; and, G.N. Artificial intelligence in urban science: why does it matter? Annals of GIS 2025, 31, 181–189. 667

49. Demuth, S.; Paris, J.; Faddeenkov, I.; De Sèze, J.; Gourraud, P.A. Clinical applications of deep learning in neuroinflammatory diseases: A scoping review. Revue Neurologique 2025, 181, 135–155.

50. López Joya, S.; Diaz-Garcia, J.; Ruiz, M.; Martin-Bautista, M. Dissecting a social bot powered by generative AI: anatomy, new trends and challenges. Social Network Analysis and Mining 2025, 15.

51. Abbas, K. Management accounting and artificial intelligence: A comprehensive literature review and recommendations for future research. The British Accounting Review 2025, p. 101551.

52. Non, L.R.; Marra, A.R.; Ince, D. Rise of the Machines - Artificial Intelligence in Healthcare Epidemiology. Current Infectious Disease Reports 2025, 27, 4.

53. Westphal, A.; Mrowka, R. Special issue European Journal of Physiology: Artificial intelligence in the field of physiology and medicine. Pflügers Archiv - European Journal of Physiology 2025, 477, 509–512.

54. Mikołajewska, E.; Mikołajewski, D.; Mikołajczyk, T.; Paczkowski, T. Generative AI in AI-Based Digital Twins for Fault Diagnosis 678

for Predictive Maintenance in Industry 4.0/5.0. Applied Sciences 2025, 15.

55. Sun, J.; Liao, V.; Muller, M.; Agarwal, M.; Houde, S.; Talamadupula, K.; Weisz, J. Investigating Explainability of Generative AI for Code through Scenario-based Design. 03 2022, pp. 212–228. https://doi.org/10.1145/3490099.3511119.

56. El-Zanfaly, D.; Huang, Y.; Dong, Y. Sand-in-the-loop: Investigating embodied co-creation for shared understandings of generative AI. 07 2023, pp. 256–260. https://doi.org/10.1145/3563703.3596652.

57. Ezzahed, Z.; Chevrot, A.; Hurter, C.; Olive, X. Bringing Explainability to Autoencoding Neural Networks Encoding Aircraft Trajectories. In Proceedings of the 13th SESAR Innovation Days 2023, SIDS 2023, Séville, Spain, 2023; pp. ISSN : 0770–1268.

58. Wang, Y.; Shen, S.; Lim, B.Y. RePrompt: Automatic Prompt Editing to Refine AI-Generative Art Towards Precise Expressions. In Proceedings of the Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, New York, NY, USA, 2023; CHI ’23. https://doi.org/10.1145/3544548.3581402. 68828 of 30

59. Jeong, S.; Li, M.; Berger, M.; Liu, S. Concept Lens: Visually Analyzing the Consistency of Semantic Manipulation in GANs. In Proceedings of the 2023 IEEE Visualization and Visual Analytics (VIS). IEEE, 2023, p. 221–225. https://doi.org/10.1109/vis54172 .2023.00053.

60. Hasko, R.; Hasko, O.; Kutucu, H. Teaching Assistant Robots in Various Fields: Natural Sciences, Medicine and Specific Non- Deterministic Conditions. In Proceedings of the 6th International Conference on Informatics and Data-Driven Medicine (IDDM 2023), Bratislava, Slovakia, 2023; Vol. 3609, pp. 303–309.

89. Hu, Y.; Giacaman, N.; Donald, C. Enhancing Trust in Generative AI: Investigating Explainability of LLMs to Analyse Confusion

in MOOC Discussions. 03 2024.

90. Toth, G.; Albrecht, R.; Pruski, C. Explainable AI, LLM, and digitized archival cultural heritage: a case study of the Grand Ducal

Archive of the Medici. AI & SOCIETY 2025, pp. 1–13. https://doi.org/10.1007/s00146-025-02238-5.

91. Di Lodovico, C.; Torrielli, F.; Di Caro, L.; Rapp, A. How Do People Develop Folk Theories of Generative AI Text-to-Image

Models? A Qualitative Study on How People Strive to Explain and Make Sense of GenAI. International Journal of Human-Computer

Interaction 2025, pp. 1–25. https://doi.org/10.1080/10447318.2025.2491009.

92. Leimeister, J.M.; Reinhard, P.; Li, M.; Fina, M. Fact or Fiction? Exploring Explanations to Identify Factual Confabulations in

RAG-Based LLM Systems. 04 2025. https://doi.org/10.1145/3706599.3720249.

93. Jeck, J.; Leiser, F.; Hüsges, A.; Sunyaev, A. TELL-ME: Toward Personalized Explanations of Large Language Models. In

Proceedings of the Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, New

York, NY, USA, 2025; CHI EA ’25. https://doi.org/10.1145/3706599.3719982.

94. Basaran, O.T.; Dressler, F. XAInomaly: Explainable and interpretable Deep Contractive Autoencoder for O-RAN traffic anomaly

detection. Computer Networks 2025, 261, 111145. https://doi.org/https://doi.org/10.1016/j.comnet.2025.111145.

95. Ahmed, M.U.; Begum, S.; Barua, S.; Masud, A.N.; Di Flumeri, G.; Navarin, N. Enhancing Explainability, Robustness, and

Autonomy: A Comprehensive Approach in Trustworthy AI. In Proceedings of the 2025 IEEE Symposium on Trustworthy,

Explainable and Responsible Computational Intelligence (CITREx). IEEE, 2025, pp. 1–7. https://doi.org/10.1109/CITREx64975

.2025.10974944.

96. Bhattacharya, A.; Stumpf, S.; De Croon, R.; Verbert, K. Explanatory Debiasing: Involving Domain Experts in the Data Generation

Process to Mitigate Representation Bias in AI Systems. In Proceedings of the Proceedings of the 2025 CHI Conference on Human

Factors in Computing Systems, New York, NY, USA, 2025; CHI ’25. https://doi.org/10.1145/3706598.3713497.

97. Katsuragi, M.; Tanaka, K. Comparing AI-Generated and Human-Crafted T-Shirt Layouts through an XAI Lens: Key Design

Elements and Implications for Co-Creative Tools. In Proceedings of the Proceedings of the Extended Abstracts of the CHI

Conference on Human Factors in Computing Systems, New York, NY, USA, 2025; CHI EA ’25. https://doi.org/10.1145/3706599.

3720016.

98. Yoshioka, T.; Morikura, Y.; Izumi, T.; Wada, T. Sustainable data-driven framework and policy recommendations for enhancing

sports promotion using generative and explainable Artificial Intelligence. Journal of Physical Education and Sport 2025, 25, 638–645.

https://doi.org/10.7752/jpes.2025.03068.

 

 

 

Generative AI Disruption Anthropomorphic Educational Robotics Lab

 

Daniel Zviel Girshin

Ruppin Knowledge Engineering AI Robotics Lab

Technical Report

Summary

 

 

Introduction

The concept of technological disruption, particularly within technological and educational contexts, has emerged as a transformative approach redefining traditional methods and structures. In this context, disruption is viewed positively, associated with optimism, enthusiasm, and creativity, driving innovation through self-confidence and emotional commitment. This paradigm advocates for a balance that favors creativity, brainstorming, and experimentation over excessive caution and skepticism, positioning itself as a revolutionary approach capable of reimagining established paradigms across various sectors, particularly in computing and education.

Central to understanding the disruption paradigm is recognizing its underlying conceptual foundations, which encourage breaking free from conventional paradigms. Traditional educational and technological environments, characterized by text-heavy interfaces and rigid instructional methods, are replaced with intuitive, playful, and boundary-pushing interactions. Inspiration for such approaches is often drawn from diverse and non-traditional sources such as gaming, storytelling, visual arts, and music, effectively shifting the mission of technology and education from mere problem-solving toward expressive and creative engagements. By embracing these influences, disruption fosters environments where young developers and learners perceive technology, particularly coding, not as an isolated technical skill, but as a means of personal and creative expression.

The disruption paradigm also emphasizes specific attributes and characteristics essential for effective implementation. Among these is a focus on user experience (UX) designed explicitly to enhance learner engagement and interaction. This includes the adoption of visual-first interfaces, gamified learning environments, and innovative interaction modalities such as virtual reality (VR), augmented reality (AR), and gesture controls. Such approaches enable users, particularly young learners, to actively participate in immersive and collaborative environments reminiscent of multiplayer gaming, facilitating greater interaction, engagement, and shared learning experiences. The integration of real-time collaborative coding environments further exemplifies this attribute, encouraging collective creativity and peer learning.

Generative Artificial Intelligence (GenAI) significantly enhances the creative process by supporting idea generation, problem-solving, debugging, and optimizing code in interactive and engaging ways. The disruption approach prioritizes the development of language-agnostic platforms that abstract complex coding languages, thereby making technology accessible to users with varying levels of technical proficiency. Furthermore, disruption encourages open-ended outputs and direct integration of projects with interdisciplinary focus that further fuels creativity and ensures technology and education remain relevant and engaging to a broad audience.

Philosophically and culturally, disruption is deeply rooted in principles of inclusivity and accessibility. By removing traditional barriers, particularly economic and technical, disruption seeks to democratize technology and education, ensuring tools and platforms are widely accessible, even on low-cost devices. Crucially, youth-first leadership is integral to the disruption paradigm, involving young individuals not merely as passive users but as active contributors and co-creators of technological and educational environments. This approach ensures authenticity, relevance, and appeal, capturing the perspectives and preferences of younger generations effectively. Additionally, disruption's cultural positioning often includes nonconformist branding, celebrating innovation, rebellion against convention, and the empowerment of young, independent thinkers who challenge traditional norms.

The historical context and evolution of the disruption paradigm further underscore its significance. Many individuals, movements, and institutions have historically advocated disruptive, revolutionary, and nonconformist approaches to technology and education. Early pioneers like Dewey, Montessori, Vygotsky and Seymour Papert exemplified these ideals by envisioning computers as creative mediums and advocating innovative educational methodologies. Revolutionary educators and designers, including Mitchel Resnick with platforms like Scratch, further expanded these disruptive ideals, demonstrating technology's profound potential to impact educational accessibility and equity.

Modern innovators, including prominent figures such as Elon Musk and Steve Jobs, have continued to advance youth-driven disruptive approaches, inspiring new generations of creative technologists. Movements such as makerspaces, DIY communities, indie game developers, and platforms like Roblox further exemplify contemporary manifestations of the disruption paradigm. These communities effectively democratize technology and foster environments where young people can actively participate, innovate, and gain entrepreneurial skills. Cultural and artistic advocates such as Douglas Rushkoff and groups inspired by Ada Lovelace have also played crucial roles, emphasizing interdisciplinary approaches and encouraging young learners to actively engage with, rather than passively consume, technology.

The operationalization of disruption within institutional frameworks, exemplified by initiatives such as the Disruption Lab, provides tangible examples of these principles in action. Such labs prioritize youth-led innovation, creativity, collaboration, and inclusivity, providing spaces and resources specifically designed to foster experimentation and interdisciplinary exploration. Physical environments within disruption labs typically feature modular zones, maker spaces, AI development sandboxes, and collaborative lounges designed to facilitate creative and practical engagements. The operational models of these labs emphasize openness, accessibility, mentorship, and user-driven project selection, significantly enhancing the practical application and effectiveness of disruptive methodologies.

Additionally, disruption labs implement structured life cycles encompassing ideation, infrastructure setup, recruitment, project launches, growth, and scaling. Such structures ensure continuous evolution and responsiveness to changing needs, emphasizing adaptability, rapid prototyping, and continuous feedback loops. Specific projects, such as anthropomorphic robots and educational robots, exemplify how disruptive principles translate into practical, real-world applications. These robots leverage multimodal communication, adaptive learning algorithms, and gamified interactions to personalize education and significantly enhance learner engagement, demonstrating disruption's potential to revolutionize educational experiences.

In summary, the disruption paradigm represents a profoundly transformative approach that redefines traditional technological and educational methodologies. It emphasizes creativity, inclusivity, collaboration, and innovation, driven by historical advocacy, contemporary practices, and future-oriented technological developments. By operationalizing these principles through structured labs, innovative projects, and interdisciplinary approaches, disruption effectively addresses the evolving needs and expectations of younger generations, ensuring technology and education remain dynamic, engaging, and profoundly impactful.