pith. sign in

arxiv: 2605.20404 · v1 · pith:WLYGV6EMnew · submitted 2026-05-19 · 💻 cs.CL

Puzzled By ChatGPT? No more! A Jigsaw Puzzle to Promote AI Literacy and Awareness

Francesca Padovani , Malvina Nissim This is my paper

Pith reviewed 2026-05-21 07:22 UTC · model grok-4.3

classification 💻 cs.CL
keywords AI literacyjigsaw puzzlegenerative AIChatGPTcomic infographicinformal learningvisual storytellingcollaborative creation
0
0 comments X

The pith

A jigsaw puzzle with comic panels teaches how ChatGPT and similar AI systems work, what they can and cannot do, and their wider effects on society.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to give the public an accessible way to grasp generative AI by turning explanations into a physical jigsaw puzzle whose finished picture forms a comic infographic. The puzzle pieces were drawn during a live session that mixed expert knowledge with everyday observations from a mixed group of participants. Once assembled, the image and its separate comic cards let users explore AI mechanisms, strengths, limits, and social consequences through hands-on play rather than lectures. A sympathetic reader would see this as a practical way to support informal learning about rapidly adopted technologies whose inner workings and risks are often opaque.

Core claim

We introduce a game-based, interactive approach in the form of a jigsaw puzzle whose completed image is a comic-based infographic illustrating the workings, capabilities, limitations, and societal implications of these technologies. Each comic sketch also functions as a standalone informational card, providing focused explanations of specific facets of AI use, design, and impact. The visual content was created in a live collaborative session with a professional illustrator and a multidisciplinary group of experts and non-experts, combining structured knowledge with informal, exploratory reflections shared during the discussion.

What carries the argument

The jigsaw puzzle whose assembled image forms a comic infographic, with each sketch acting as both a puzzle piece and a self-contained informational card.

If this is right

  • Users gain concrete knowledge of AI workings by physically connecting comic panels that show both internal processes and external effects.
  • The standalone comic cards allow focused discussion of individual topics such as capabilities, limits, or societal risks without needing the full puzzle.
  • Informal group settings become sites for exploring both the benefits and the drawbacks of generative AI through shared assembly and conversation.
  • The method supplies a low-cost, repeatable activity that can be used in classrooms, community events, or homes to build basic AI awareness.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar puzzles could be created for other emerging technologies whose workings are hard to visualize from text alone.
  • The collaborative creation step may itself serve as a model for how experts and non-experts can jointly produce public-facing explanations.
  • Repeated use of the puzzle in different groups might reveal which aspects of AI are most easily misunderstood and therefore most in need of visual treatment.
  • If the puzzle format proves engaging, it could be adapted into digital versions or classroom kits for wider distribution.

Load-bearing premise

The explanations produced during the single live collaborative session are accurate and accessible enough to support genuine AI literacy once turned into a puzzle.

What would settle it

If people who assemble and discuss the puzzle show no measurable gain in understanding AI mechanisms or implications compared with people who read equivalent text explanations, the approach would fail its intended purpose.

Figures

Figures reproduced from arXiv: 2605.20404 by Francesca Padovani, Malvina Nissim.

Figure 1
Figure 1. Figure 1: The complete jigsaw puzzle image1 . I. INTRODUCTION AND MOTIVATION The increasing pervasiveness of Generative Artificial Intel￾ligence (GenAI), including Large Language Model (LLM)- based chatbots like ChatGPT [1], across social, professional, 1Patches in the bottom-right conceal institutional logos and funding ac￾knowledgements to preserve author anonymity during review. The copyright notice in the bottom… view at source ↗
Figure 2
Figure 2. Figure 2: Front/back of a card on the theme “ChatGPT is not a database”. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Front/back of a card introducing the Transformer architecture. A. Co-Creating the Puzzle Visuals: Live Collaborative Session The visual content of the puzzle was developed in a ded￾icated live session, designed to translate multifaceted notions and thoughts about conversational agents and GenAI into comics-based narratives. During this session, participants— both experts and non-experts of AI—discussed a r… view at source ↗
Figure 4
Figure 4. Figure 4: Puzzle boxes and some pieces in the background. [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Puzzle stations during the European Researchers’ Night. [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
read the original abstract

The rapid adoption of Generative AI, including LLM-based chatbots like ChatGPT, has highlighted the need for accessible ways to support public understanding and AI literacy. To address this need, we introduce a game-based, interactive approach in the form of a jigsaw puzzle whose completed image is a comic-based infographic illustrating the workings, capabilities, limitations, and societal implications of these technologies. Each comic sketch also functions as a standalone informational card, providing focused explanations of specific facets of AI use, design, and impact. The visual content was created in a live collaborative session with a professional illustrator and a multidisciplinary group of experts and non experts, combining structured knowledge with informal, exploratory reflections shared during the discussion. By integrating hands-on assembly, visual storytelling, and collaborative interaction, the puzzle provides an engaging and playful tool for exploring the mechanisms, perks, and perils of AI systems in informal learning contexts.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper introduces a jigsaw puzzle whose completed image forms a comic-based infographic on the mechanisms, capabilities, limitations, and societal implications of generative AI systems such as ChatGPT. Each panel doubles as a standalone informational card. The visual content was produced in a live collaborative session involving a professional illustrator and a multidisciplinary group of experts and non-experts; the authors claim that hands-on assembly combined with visual storytelling yields an engaging tool for AI literacy in informal learning settings.

Significance. If shown to be effective, the work would supply a concrete, low-cost physical resource that combines tactile interaction with accessible visual explanations, potentially filling a gap in public AI education where text-based or screen-only materials fall short. The collaborative creation method itself illustrates one practical route for translating technical concepts into broadly understandable form.

major comments (2)
  1. [Abstract] Abstract: the central claim that the puzzle 'provides an engaging and playful tool for exploring the mechanisms, perks, and perils of AI systems in informal learning contexts' is asserted without any user testing, pre/post knowledge measures, comprehension checks, or comparative evaluation; the manuscript reports only the design and creation process.
  2. [Creation process section] Creation-process description: the assumption that a single live session with experts and non-experts automatically yields accurate, accessible explanations rests on the authors' participation alone; no verification steps, external review of content accuracy, or discussion of how informal reflections were reconciled with established technical facts are provided.
minor comments (1)
  1. [Abstract] The manuscript would benefit from an explicit statement of the intended age range or prior-knowledge level of the target users.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive comments. We appreciate the positive assessment of the work's potential significance and have revised the manuscript to address the concerns regarding unsubstantiated claims and the description of the creation process.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the puzzle 'provides an engaging and playful tool for exploring the mechanisms, perks, and perils of AI systems in informal learning contexts' is asserted without any user testing, pre/post knowledge measures, comprehension checks, or comparative evaluation; the manuscript reports only the design and creation process.

    Authors: We acknowledge that the manuscript primarily reports on the design and creation process of the jigsaw puzzle without including user testing or evaluation measures. The central claim in the abstract was intended to describe the intended purpose and design rationale rather than an empirically validated outcome. To correct this, we will revise the abstract to state that the puzzle is designed to provide an engaging tool, based on the collaborative development process, while removing any implication of demonstrated effectiveness. This revision ensures the claims are supported by the content presented. revision: yes

  2. Referee: [Creation process section] Creation-process description: the assumption that a single live session with experts and non-experts automatically yields accurate, accessible explanations rests on the authors' participation alone; no verification steps, external review of content accuracy, or discussion of how informal reflections were reconciled with established technical facts are provided.

    Authors: We agree that the original description could benefit from more detail on ensuring accuracy. The live session incorporated input from experts in AI and related fields to ground the content in technical facts, with informal reflections from non-experts helping to identify accessible explanations. However, no formal external review or structured verification process was conducted beyond the session itself. We will revise the creation process section to explicitly describe the sources of technical accuracy (e.g., drawing from established AI literature and expert consensus during the session) and discuss the reconciliation of reflections with facts. This will clarify the process without overstating its rigor. revision: yes

Circularity Check

0 steps flagged

No circularity: purely descriptive account of artifact creation

full rationale

The manuscript is a descriptive presentation of a jigsaw puzzle's collaborative development and component design for AI literacy. It contains no derivations, equations, fitted parameters, predictions, or modeling steps that could reduce to inputs by construction. The central claim rests on narrative description of the live session and comic panels rather than any self-referential logic, self-citation chain, or renamed empirical pattern. No load-bearing step invokes uniqueness theorems or ansatzes from prior work by the authors. This is a standard non-circular descriptive paper on an educational artifact.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is a design-description paper rather than a theoretical or empirical study, so the central claim rests on no free parameters, mathematical axioms, or invented entities.

pith-pipeline@v0.9.0 · 5686 in / 1106 out tokens · 51558 ms · 2026-05-21T07:22:48.114951+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages · 1 internal anchor

  1. [1]

    ChatGPT,

    OpenAI, “ChatGPT,” https://chat.openai.com, 2023

    work page 2023

  2. [2]

    AI-GO! Een Raamwerk voor AI-Geletterdheid in het Onderwijs (AI-GO Raamwerk),

    M. Renkema, E. van den Boom-Muilenburg, I. Friso-van den Bos, H. Theelen, I. Wopereis, and K. Schildkamp, “AI-GO! Een Raamwerk voor AI-Geletterdheid in het Onderwijs (AI-GO Raamwerk),” Npuls, Tech. Rep., 2025

    work page 2025

  3. [3]

    Metacognitive AI literacy: findings from an interactive AI fair,

    B. Hutler, Y . Alzahrani, A. Olagunju, K. Doherty, M. Tallapragada, and B. W. Hardy, “Metacognitive AI literacy: findings from an interactive AI fair,”AI & SOCIETY, pp. 1–16, 2025

    work page 2025

  4. [4]

    Most visited websites in the world (Au- gust 2025),

    Exploding Topics, “Most visited websites in the world (Au- gust 2025),” [Online]. Available: https://explodingtopics.com/blog/ most-visited-websites, 2025, accessed: Mar. 15, 2026

    work page 2025

  5. [5]

    Artificial intelligence across Europe: A study on awareness, attitude and trust,

    T. Scantamburlo, A. Cort ´es, F. Foffano, C. Barru ´e, V . Distefano, L. Pham, and A. Fabris, “Artificial intelligence across Europe: A study on awareness, attitude and trust,”IEEE Transactions on Artificial Intelligence, vol. 6, no. 2, pp. 477–490, 2024

    work page 2024

  6. [6]

    Generative AI Practices, Literacy, and Divides: An Empirical Analysis in the Italian Context

    B. Savoldi, G. Attanasio, O. Gorodetskaya, M. M. Manerba, E. Bassig- nana, S. Casola, M. Negri, T. Caselli, L. Bentivogli, A. Ramponiet al., “Generative AI Practices, Literacy, and Divides: An Empirical Analysis in the Italian Context,”arXiv preprint arXiv:2512.03671, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  7. [7]

    Generative AI and gamification for personalized learning: Literature review and future challenges,

    F. Abbes, S. Bennani, and A. Maalel, “Generative AI and gamification for personalized learning: Literature review and future challenges,”SN Computer Science, vol. 5, no. 8, p. 1154, 2024

    work page 2024

  8. [8]

    Why do people use ChatGPT? Exploring user motivations for generative conversational AI,

    M. Skjuve, P. B. Brandtzaeg, and A. Følstad, “Why do people use ChatGPT? Exploring user motivations for generative conversational AI,” First Monday, 2024

    work page 2024

  9. [9]

    Experimental evidence on the productivity effects of generative artificial intelligence,

    S. Noy and W. Zhang, “Experimental evidence on the productivity effects of generative artificial intelligence,”Science, vol. 381, no. 6654, pp. 187–192, 2023

    work page 2023

  10. [10]

    Generative AI at work,

    E. Brynjolfsson, D. Li, and L. Raymond, “Generative AI at work,”The Quarterly Journal of Economics, vol. 140, no. 2, pp. 889–942, 2025

    work page 2025

  11. [11]

    Deep play: Notes on the balinese cockfight,

    C. Geertz, “Deep play: Notes on the balinese cockfight,”Daedalus, vol. 134, no. 4, pp. 56–86, 2005

    work page 2005

  12. [12]

    Adult play-learning: Observing informal family education at a science museum,

    P. Kanhadilok and M. Watts, “Adult play-learning: Observing informal family education at a science museum,”Studies in the Education of Adults, vol. 46, no. 1, pp. 23–41, 2014

    work page 2014

  13. [13]

    Considering play: From method to analysis,

    J. Van Vught and R. Glas, “Considering play: From method to analysis,” Transactions of the Digital Games Research Association, vol. 4, no. 2, 2018

    work page 2018

  14. [14]

    Play as a Method: Positionality, Power, Possibility,

    D. Leorke, “Play as a Method: Positionality, Power, Possibility,”Cultural Studies↔Critical Methodologies, vol. 26, no. 2, pp. 131–141, 2026

    work page 2026

  15. [15]

    Play Studies: A Brief History

    T. S. Henricks, “Play Studies: A Brief History.”American Journal of Play, vol. 12, no. 2, pp. 114–155, 2020

    work page 2020

  16. [16]

    Homo ludens 2.0: Play, media, and identity,

    V . Frissen, S. Lammes, M. De Lange, J. De Mul, and J. Raessens, “Homo ludens 2.0: Play, media, and identity,” inPlayful identities. Routledge, 2025, pp. 9–50

    work page 2025

  17. [17]

    An explorer in a cardboard land: emotion, memory, and the embodied experience of doing jigsaw puzzles,

    A. C. Garcia, “An explorer in a cardboard land: emotion, memory, and the embodied experience of doing jigsaw puzzles,”International Journal of Play, vol. 5, no. 2, pp. 166–180, 2016

    work page 2016

  18. [18]

    A. D. Williams,The jigsaw puzzle: Piecing together a history. Berkley Hardcover, 2004

    work page 2004

  19. [19]

    Having fun and accepting challenges are natural instincts: jigsaw puzzles to challenge students and test their abilities while having fun!

    H. R. Rodenbaugh, H. L. Lujan, D. W. Rodenbaugh, and S. E. DiCarlo, “Having fun and accepting challenges are natural instincts: jigsaw puzzles to challenge students and test their abilities while having fun!” Advances in physiology education, vol. 38, no. 2, pp. 185–186, 2014

    work page 2014

  20. [20]

    Transforming Language Education Through Jigsaw Puzzle: A Focus on Enhancing Learning Continuity and Recovery Amidst Educational Crisis,

    J. B. L. Layoc and M. A. D. Elepa ˜no, “Transforming Language Education Through Jigsaw Puzzle: A Focus on Enhancing Learning Continuity and Recovery Amidst Educational Crisis,” inInt’l Conf on Frontiers of Educational Technologies. Springer, 2025, pp. 87–98

    work page 2025

  21. [21]

    Alteration of time perception in young and elderly people during jigsaw puzzle tasks with different complexities,

    Y . Iwamoto and M. Hoshiyama, “Alteration of time perception in young and elderly people during jigsaw puzzle tasks with different complexities,”Occupational therapy international, vol. 18, no. 4, pp. 194–200, 2011

    work page 2011