This paper presents an eXplainable AI (XAI)-based classroom game “Breakable Machine” for teaching critical, transformative AI literacy through adversarial play and interrogation of AI systems. Designed for learners aged 10–15, the game invites students to spoof an image classifier by manipulating their appearance or environment in order to trigger high-confidence misclassifications. Rather than focusing on building AI models, this activity centers on breaking them—exposing their brittleness, bias, and vulnerability through hands-on, embodied experimentation. The game includes an XAI view to help students visualize feature saliency, revealing how models attend to specific visual cues. A shared classroom leaderboard fosters collaborative inquiry and comparison of strategies, turning the classroom into a site for collective sensemaking. This approach repositions AI education by treating model failure and misclassification not as problems to be debugged, but as pedagogically rich opportunities to interrogate AI as a sociotechnical system. In doing so, the game supports students in developing data agency, ethical awareness, and a critical stance toward AI systems increasingly embedded in everyday life.

In the fast-growing field of K–12 AI education, there is an urgent need for accessible, hands-on tools that introduce AI concepts and workflows to novice learners. In recent years, a variety of AI education tools have been introduced, ranging from coding environments to physical kits and robots. To provide an alternative to existing AI education tools, this paper presents a low-cost robotics kit (

This paper presents an explainable AI (XAI) education tool designed for K-12 classrooms, particularly for students aged 11-16. The tool was designed for interventions on the fundamental processes behind social media platforms, focusing on four AI- and data-driven core concepts: data collection, user profiling, engagement metrics, and recommendation algorithms. An Instagram-like interface and a monitoring tool for explaining the data-driven processes make these complex ideas accessible and engaging for young learners. The tool provides hands-on experiments and real-time visualizations, illustrating how user actions influence their personal experience on the platform as well as the experience of others. This approach seeks to enhance learners' data agency, AI literacy, and sensitivity to AI ethics. The paper includes a case example from 12 two-hour test sessions involving 209 children, using learning analytics to demonstrate how they navigated their social media feeds and the browsing patterns that emerged.

This study investigates the rarely-explored scaffolding processes in teaching artificial intelligence (AI), more specifically machine learning (ML), to K-12 students using educational technology. Focusing on a ML workshop within a children’s science camp, we observed 7-12 year-olds interacting with image classifiers using their own drawings, guided by an experienced computing teacher. Our analysis highlights the importance of the teacher’s role in a technology-rich environment in using diagnostic questions to reveal and address students’ misconceptions, aligning with the concept of contingent support. By linking theoretical concepts to practical activities, the teacher helped shift the focus from surface features to deeper processes, promoting advanced reasoning. The paper discusses a distributed scaffolding system combining teacher guidance, technological affordances, and peer interaction, crucial for making complex concepts accessible to young learners. These insights are important for educators and technology developers in enhancing K-12 ML/AI education.

This paper introduces an educational tool for classroom use, based on explainable AI (XAI), designed to demystify key social media mechanisms—tracking, profiling, and content recommendation—for novice learners. The tool provides a familiar, interactive interface that resonates with learners’ experiences with popular social media platforms, while also offering the means to “peek under the hood” and exposing basic mechanisms of datafication. Learners gain first-hand experience of how even the slightest actions, such as pausing to view content, are captured and recorded in their digital footprint, and further distilled into a personal profile. The tool uses real-time visualizations and verbal explanations to create a sense of immediacy: each time the user acts, the resulting changes in their engagement history and their profile are displayed in a visually engaging and understandable manner. This paper discusses the potential of XAI and educational technology in transforming data and digital literacy education and in fostering the growth of children’s privacy and security mindsets.

This paper introduces an AI education tool designed for novice learners to create machine learning (classifier) based applications. Advancing from Google’s Teachable Machine 2 and developed using the design science research methodology, the tool is piloted in 36 K-12 classroom sessions with 213 children and allows learners to easily navigate the complete ML workflow—from data collection to app deployment—without any programming skills. To evaluate how well the tool met children’s expectations children were asked, as part of the design process, to articulate their goals and intentions for their apps; then, after using the tool, to describe how well they perceived their final app realized their intention. The tool’s main novelty is its ability to create a standalone app by defining one or more actions to be triggered by each classifier result, and deploy that app to other devices. A no-code approach and fully integrated development environment reduces the need for technical skills, making AI learning more inclusive. The tool represents a significant step in making AI education accessible for early learners, with future enhancements aimed at expanding its capabilities.

K-12 computing education research is a rapidly growing field of research, both driven by and driving the implementation of computing as a school and extra-curricular subject globally. Within discipline-based education research, it is a new and emerging field, drawing on fields such as mathematics and science education research for inspiration and theoretical bases. The urgency around investigating effective teaching and learning in computing in school alongside broadening participation has led to much of the field being focused on empirical research. Less attention has been paid to the underlying philosophical assumptions informing the discipline, which might include a critical examination of the rationale for K-12 computing education, its goals and perspectives, and associated inherent values and beliefs. The goals of this research project are to understand the implicit and hidden values, perspectives and goals underpinning computing education at school. This will be achieved through a critical examination of a wide body of literature leading to the development of a categorization and framework.

Generative AI is a recent breakthrough in AI. While it has become a hot topic in computing education research (CER), much of the recent research has focused on e.g. issues of plagiarism or academic integrity. One problem spot with Generative AI is its susceptibility to various kinds of algorithmic bias. In this study, we collected data from an introductory computing course, where students experimented with text-to-image generative models and reflected on their generated image sets, in terms of biases, related harms, and possible fixes. Data were collected in Fall 2023 (pilot data in Fall 2022). Data included reports from 163 students. The results show (1) a variety of bias types observed by students related to gender, ethnicity, age, as well as a variety of bias types not observed by students, (2) two major types of attributions for the source of bias: bias caused by biases in the society and bias caused by data or algorithms, and (3) a number of potential harms associated with the biases, as well as attributions of those harms in specific contexts and use cases.

In this poster, we present the results from the co-design school projects on machine learning. We address social and educational challenges in artificial intelligence including security, privacy and education. We employ the participatory co-design approach, which facilitates children's right to be heard, and positions them as active partners, advisers, and designers in research and development work on technology and socio-technological practices.

The question of how to teach classical, rule-based programming has been driving much of the computing education research since the 1950s. In the K--12 (school) context, a consensus has emerged over time on the paradigmatic elements of computing education, which implicitly assumes a von Neumann computer executing instruction sequences guided by imperative programs. Within this framework, many researchers have focused on how to facilitate learners to develop an accurate mental model of what the computer does when it executes a piece of code.

This Research Full Paper addresses the definition and implementation of Computational Thinking (CT) in K-12 education. CT is the focus of ongoing debate about the future of computing in schools, and this poses a number of challenges. Early discussions on the topic were plagued by vague and all-encompassing definitions of the term, which raised awareness of a need for school learning focused on emerging computational systems and their impact on society and the lives of citizens, but failed to address what should actually be included in the revised curricula. As some of those issues have been addressed, a number of problems remain. This paper analyses some recent and future developments in the field of computing, and expands the discussion on the vexed question of the nature and limits of computational thinking. We ask whether the current mainstream literature and curriculum for CT is well positioned in terms of the future towards which computing is heading. We conclude by summarising the implications these questions have for the future development of K-12 curricula and teaching practices.

This innovative practice full paper is grounded in the societal developments of computing in the 2000s, which have brought the concept of information literacy and its many variants into limelight. Widespread tracking, profiling, and behavior engineering have set the alarms off, and there are increasing calls for education that can prepare citizens to cope with the latest technological changes. We describe an active concept, data agency, that refers to people's volition and capacity for informed actions that make a difference in their digital world. Data agency extends the concept of data literacy by emphasizing people's ability to not only understand data, but also to actively control and manipulate information flows and to use them wisely and ethically.This article describes the theoretical underpinnings of the data agency concept. It discusses the epistemological and methodological changes driven by data-intensive analysis and machine learning. Epistemologically the many new modalities of automation are non-reductionist, non-deterministic, and statistical; the models they rely on are soft and brittle. This article also presents results from a pilot study on how to teach central machine learning concepts and workflows in K-12 through co-creation of machine learning-based solutions.

This Research to Innovative Practice Full Paper presents a multidisciplinary, design-based research study that aims to develop and study pedagogical models and tools for integrating machine-learning (ML) topics into education. Although children grow up with ML systems, few theoretical or empirical studies have focused on investigating ML and data-driven design in K-12 education to date. This paper presents the theoretical grounds for a design-oriented pedagogy and the results from exploring and implementing those theoretical ideas in practice through a case study conducted in Finland. We describe the overall process in which middle-schoolers (N = 34) co-designed and made ML applications for solving meaningful, everyday problems. The qualitative content analysis of the pre-and post-tests, student interviews, and the students' own ML design ideas indicated that co-designing real-life applications lowered the barriers for participating in some of the core practices of computer science. It also supported children in exploring abstract ML concepts and workflows in a highly personalized and embodied way. The article concludes with a discussion on pedagogical insights for supporting middle-schoolers in becoming innovators and software designers in the age of ML.

Artificial intelligence and machine learning are making their ways rapidly to K-12 education. Google Teachable Machine, powered by convolutional neural networks, provides an easy-to-use yet powerful tool for classification tasks. We conducted a series of co-design workshops with primary school children, where they explored and designed their own machine learning powered applications with Google Teachable Machine. Our results show that Google Teachable Machine is a feasible tool for K-12 education. The trained machine learning models are lightweight and computationally efficient, and the applications are usable even with low-end mobile devices. The students and teachers appreciated the multidisciplinary and inclusive workshop, which supports development of transversal competencies in accordance to the national primary school curriculum.