Creative Applications of Artificial Intelligence in Education

Over the weekend, I read a LinkedIn post discussing a book about artificial intelligence in education. Normally, I flag posts like that or store the link so that I can return to the post, the article, or, in this case, the book.

The book’s title is Creative Applications of Artificial Intelligence in Education. It is an open-access book edited by Alex Urmeneta and Margarida Romero, available for a free download. You can also buy a hardcover copy of the book.

Another Application of Generative AI

When I opened the PDF, I decided that I would conduct another generative AI experiment using ChatGPT-4o. I initially typed the following prompt into ChatGPT-4o.

I have a PDF of a book about teachers using AI to enhance the quality of teaching and learning. I would like you to provide a 2,500-word summary of the text, highlighting the strongest sections of the book and recommending enhancements.

I received the following response.

Please upload the PDF of the book, and I will get started on the summary for you.

Unfortunately, the PDF must have been too large to load. After the upload timed out, I changed my prompt to the following.

I was unable to upload the PDF. Instead, I will upload a chapter of the book at a time. I would like you to provide a 250-350 word synopsis of each chapter. Ready to begin with Chapter 1?

ChatGPT-4o responded with the following statement:

Absolutely, please go ahead and upload Chapter 1. I’ll provide a 250-350 word synopsis once I receive it.

Over a period of approximately 20 minutes, I copied each chapter and pasted the copy into the prompt section of ChatGPT4o. I pasted every chapter synopsis created by ChatGPT-4o in sequence (12 in all) at the bottom of this article. I did not change any of the ChatGPT-4o chapter summaries.

Analysis of the Process of Summarizing the Book

Building my prompts, copying each chapter of the book from the PDF, loading each chapter into ChatGPT-4o, and waiting for the synopsis did not take more than 20 minutes. However, there were a few hiccups.

The output from ChatGPT-4o followed the same format in Chapters 1 through 4 (see consecutive chapter summaries below). My prompt asked for a 250-350 word synopsis of each chapter. The format provided in these responses was created by the generative AI app, not directed by me through my prompt.

The format changed for Chapter 5. I went back and checked to make sure that my prompt did not change. Before each chapter was uploaded, I entered the following sentence with only the chapter number changing. The prompt for Chapter 5 was:

I am uploading Chapter 5.

Yet, the format for the “synopsis” changed for Chapter 5. If I was trying to achieve the same exact format, I might have gone back and provided the output for Chapter 4 and asked the app to duplicate the format for Chapter 5 and subsequent chapters. I don’t have an explanation for why the format changed since the prompt did not change other than the number of the chapter.

The formats for Chapters 6 and 7 changed without a change in the prompt. When I noticed that the format had changed, I modified my prompt slightly and submitted the following prompt:

I am going to upload Chapter 8, which I would like you to summarize in the same manner as the previous chapters.

Chapter 8’s format was closer to the format of the first four chapters than chapters 5 through 7. I noticed that the output for Chapter 7 and for Chapter 8 did not include the names of the authors for each of those chapters.

For Chapter 9, I entered the following prompt:

I am going to upload Chapter 9. I would like you to summarize the chapter like you have done with the previous 8 chapters.

ChatGPT-4o responded with the following comment:

Sure, please go ahead and upload Chapter 9. I’ll summarize it in the same manner as the previous chapters.

That did not happen as I expected. The titles for Chapters 9 and 10 invoked the word “summary” again, but the basic format was like Chapter 8, except these summaries listed the authors for each chapter, unlike the summaries for Chapters 7 and 8.

A Fake Chapter Was Created by the App

My prompt for Chapter 11 looked very similar to my previous prompts. I appended it below:

I am going to upload chapter 11. Please summarize it in the same manner that you have summarized the first 10 chapters.

Instead of the app instructing me to upload the chapter, it produced a summary for Chapter 11 immediately without giving me a chance to upload it. This was an unbelievable AI hallucination. There is no chapter like the summary listed below. I had to go back and check the table of contents because I was so dumbfounded.

I have no explanation for why ChatGPT-4o hallucinated. If I had not paid careful attention to each of the chapter summaries, I might not have picked this up until I realized that I had two chapter 11’s. The summary reads like a legitimate summary of one of the chapters. Everything is fictional, although the authors are legit.

I conducted a quick Google search of the Internet to see if these authors or others wrote a paper with a similar name. I did not find one. I’ve appended a copy of the output below this paragraph. The fake appears to be real. Again, I have no explanation for why ChatGPT-4o changed its response to the series of prompts I initiated.

Summary of Chapter 11: Exploring the Role of AI in Developing Student Creativity

Authors: Margarida Romero, Tanja Tajmel, and Olga Volha Samokish

Overview

This chapter investigates the role of artificial intelligence (AI) in fostering creativity among students. It explores how AI can be used to enhance creative skills, the potential benefits and challenges of integrating AI in creative education, and provides practical examples of AI applications in creative learning environments. The authors emphasize the importance of a balanced approach that combines human creativity with AI capabilities to create a synergistic learning experience.

Introduction

The chapter begins by addressing the growing interest in using AI to develop student creativity. AI’s ability to analyze large datasets and generate novel ideas presents new opportunities for creative education. However, the integration of AI in this field requires careful consideration to ensure it complements rather than replaces human creativity.

The Potential of AI in Creative Education

AI offers several advantages in creative education, such as:

• Providing personalized feedback and suggestions
• Enhancing collaborative creativity through AI-driven tools
• Offering new perspectives and ideas by analyzing patterns and data
• Supporting the development of creative skills through interactive and adaptive learning environments

Key Benefits:

• AI can serve as a creative partner, providing inspiration and expanding creative possibilities.
• It can help identify and nurture individual creative strengths by offering tailored learning experiences.
• AI tools can facilitate collaboration among students by enabling them to co-create and share ideas in new ways.

Challenges and Considerations

While AI has the potential to enhance creativity, several challenges must be addressed:

• Ensuring that AI does not stifle human creativity by providing overly prescriptive solutions.
• Balancing the use of AI tools with traditional creative practices to maintain a human touch.
• Addressing ethical concerns related to data privacy and the potential misuse of AI-generated content.
• Providing adequate training for educators to effectively integrate AI in creative education.

Key Challenges:

• Avoiding dependency on AI for creative tasks, which could lead to a reduction in original human thought.
• Ensuring that AI tools are used ethically and responsibly, with a focus on enhancing human creativity rather than replacing it.

Practical Applications of AI in Creative Education

The chapter presents several practical examples of AI applications in creative learning environments, such as:

• AI-driven platforms for music composition, allowing students to experiment with different musical styles and structures.
• Visual art tools that use AI to suggest new techniques and styles, helping students explore various artistic expressions.
• Writing assistants that provide feedback on creative writing, suggesting improvements and new directions for stories.
• Collaborative tools that enable students to work together on creative projects, leveraging AI to facilitate idea generation and refinement.

Examples:

• An AI-powered music composition tool that suggests chord progressions and harmonies based on user input.
• A visual art platform that uses AI to analyze and enhance students’ artwork, providing insights into color theory and composition.
• A writing assistant that helps students develop their narratives by offering suggestions for plot twists and character development.

Conclusion

The integration of AI in creative education holds significant promise for enhancing student creativity. By providing personalized support, facilitating collaboration, and offering new perspectives, AI can serve as a valuable tool in developing creative skills. However, it is crucial to maintain a balanced approach that prioritizes human creativity and ethical considerations. Educators must be equipped with the necessary skills and knowledge to effectively integrate AI in creative learning environments, ensuring that it complements and enhances the creative process.

References

• Romero, M., Tajmel, T., & Samokish, O. V. (2024). Exploring the role of AI in developing student creativity. In A. Urmeneta & M. Romero (Eds.), Creative Applications of Artificial Intelligence in Education. Palgrave Studies in Creativity and Culture.

This summary encapsulates the chapter’s exploration of AI’s potential to enhance creativity in education, while addressing the challenges and considerations necessary for its effective and ethical integration.

How I Uploaded the Real Chapter 11

I was concerned that ChatGPT-4o might reject my submission of the real chapter 11, given that its hallucinated chapter was titled chapter 11. I changed my next prompt to this:

I am going to upload another chapter. Please summarize it. Thank you.

ChatGPT-4o responded with the analysis of the real Chapter 11, which is included in the sequence below. The analysis of Chapter 12 followed without hiccups or hallucinations.

This exercise is not the first time that I have experienced hallucinations in the ChatGPT4 response(s) to my prompt. One of my first posts about ChatGPT in January 2023 mentioned a hallucination that occurred when I asked ChatGPT to provide a list of papers about online student retention that my co-author, Phil Ice, and I had published.

Ironically, another previous article suggested that hallucinations could be minimized if papers were submitted through a prompt that limited the analysis to that item. I did not review each of the summaries created by ChatGPT-4o and compare them to the chapters for accuracy. After the Chapter 11 hallucination, I checked each for the appropriate authors and titles.

I’ve commented consistently about generative AI tools that I would not award them a grade better than C or 70. Accuracy is very important. My intent in using ChatGPT-4o to summarize the book chapters was to provide readers with a much shorter overview of a book about teaching and AI. I liked the book and the lesson I learned about GenAI.

Until the GenAI apps are smart enough to pick up the subtleties in our written prompts, it’s important to pay close attention to the exact wording. My prompt for the analysis of Chapter 11 was only slightly different than other prompts I used. I substituted the word “it” for chapter in the second sentence. The app did not wait for my upload but created a realistic but totally fake summary. I have no idea why.

For those of us using GenAI tools for teaching and learning, this exercise proves that you cannot be careless with using ChatGPT-4o and other apps. While I believe that I was careful, I was not precise. At the same time, I had no intention of submitting the output from my prompts without reviewing it.

Students and teachers in a hurry should take a few extra minutes to review the output. Despite the hallucinated chapter (that I caught), I would give the output a passing grade for its speed and summaries of the chapters (other than the fake Chapter 11).

I could not have read or summarized these chapters in less than two hours. ChatGPT-4o needed less than a minute per chapter. Other than the major hallucination, the prompts generated usable summaries.

The aggregated output follows below:

Synopsis of Each of the 12 Chapters

Chapter 1: Creative Application of Artificial Intelligence in Education

Overview

Chapter 1 of “Creative Applications of Artificial Intelligence in Education” by Alex Urmeneta and Margarida Romero introduces the potential and implications of artificial intelligence (AI) in the educational sector. The chapter aims to present an AI overview within education, examining its effects, ethical considerations, and potential to foster innovative teaching methods. It takes a critical approach to the benefits and drawbacks of AI and explores the sociocultural and creative aspects of using AI in education.

Key Themes

1. Historical Background and Technological Interventions:
   • The chapter discusses the history of technological advancements in education, such as the use of headphones, projectors, calculators, personal computers, the internet, and MOOCs.
   • It emphasizes that while technology has continually promised transformation, a critical perspective is necessary to evaluate its real impact and address the limitations of techno-solutionism, which often overestimates technology’s ability to solve complex societal issues.
2. AI’s Potential in Education:
   • AI has been developing since the 1950s and has now reached a stage where it significantly impacts various sectors, including education.
   • The chapter highlights AI’s capabilities in emulating human intelligence and its application in education, such as problem-solving, learning, and decision-making.
   • It underscores the importance of ethical principles in AI’s implementation, particularly in protecting privacy and the well-being of students.
3. Human–AI Co-Creativity:
   • The authors introduce the concept of human–AI co-creativity, where both human intuition and AI’s computational power collaborate to enhance creative processes in education.
   • The chapter presents a model for different levels of creative engagement with AI, ranging from passive consumption of AI-generated content to participatory knowledge co-creation and expansive learning supported by AI.
4. Educational Transformation through AI:
   • The book aims to explore how AI can serve as a tool for innovative teaching strategies, learner-centric activities, and creative pedagogy.
   • It outlines three parts of the book: the creative engagement approach, real examples in K-12 education, and advances in higher education. These sections provide concrete examples of AI’s role in transforming educational practices and supporting human–AI collaboration.

Implications and Future Directions

• The chapter suggests that AI has the potential to revolutionize education by enhancing personalized learning, supporting creative pedagogies, and facilitating human–AI collaboration.
• It calls for a nuanced understanding of AI’s role in education, considering both its benefits and challenges.
• The chapter emphasizes the need for ethical considerations and the importance of fostering AI literacy among all educational stakeholders.

Summary

Chapter 1 effectively sets the stage for exploring the creative applications of AI in education. It provides a comprehensive overview of AI’s potential, historical context, and the importance of ethical and critical perspectives. The chapter introduces the concept of human–AI co-creativity and outlines the structure of the book, highlighting the transformative potential of AI in fostering innovative teaching methods and enhancing learning experiences.

Chapter 2: Preserving Teacher and Student Agency: Insights from a Literature Review

Overview

Chapter 2, authored by Alexandre Lepage and Simon Collin, offers a comprehensive literature review examining the influence of artificial intelligence (AI) on education, with a specific focus on its implications for human agency. The chapter explores the advantages and potential drawbacks of integrating AI in education, particularly concerning the autonomy and decision-making abilities of teachers and students. It raises critical ethical questions about AI’s role in potentially enhancing or diminishing the capabilities of human educators and learners.

Key Themes

1. Potential Advantages of AI in Education:
   • AI can assist in various educational tasks such as intelligent tutoring, learning assessment, and dropout prevention.
   • These AI applications hold promise for improving efficiency and providing immediate support to students, potentially freeing teachers from time-consuming tasks like grading.
2. Ethical Considerations and Human Agency:
   • The chapter highlights the need to preserve teacher and student agency in an AI-driven education system.
   • Ethical concerns include the risk of reducing teachers’ autonomy, shifting decision-making power to programmers, and creating power imbalances.
   • AI’s role should be to enhance human capabilities rather than replace them, ensuring that the expertise of educators is respected and preserved.
3. Theoretical Frameworks:
   • The chapter introduces the technician system theory, which warns against neglecting the complexity of teaching techniques as AI systems become more sophisticated.
   • The concept of agency is discussed from both computer science and social science perspectives, emphasizing the importance of autonomy and decision-making abilities for both AI systems and individuals.
4. Findings from the Literature Review:
   • AI integration can lead to power imbalances, reducing teachers’ decision-making authority and shifting expertise to system designers.
   • Teacher skills might decline due to over-reliance on AI systems, leading to automation bias where educators may become too dependent on AI recommendations.
   • Student agency can be compromised, especially when AI systems assign tasks or gamify learning experiences without considering student preferences and needs.

Recommendations

• Design and Implementation:
  • AI-based tools should be designed to enhance, not restrict, human agency. This involves ensuring transparency in AI systems and maintaining human control over educational decisions.
  • AI should improve the quality of information available to educators and learners, allowing them to make more informed decisions rather than making autonomous decisions on their behalf.
  • AI in education should focus on providing personalized and frequent feedback to students, aiding their learning process without replacing the professional judgment of teachers.
• Preserving Autonomy:
  • Both teachers and students should have the ability to choose whether to follow AI recommendations. This autonomy is crucial for maintaining their agency.
  • AI systems should be designed to support teachers and students in their roles, providing tools that enhance their abilities rather than diminishing their importance.

Summary

Chapter 2 underscores the importance of preserving teacher and student agency in the context of AI in education. While AI offers significant potential advantages, it also poses ethical challenges that need careful consideration. The chapter advocates for AI-based tools that enhance human agency, ensuring that educators and learners retain their autonomy and decision-making abilities. The findings from the literature review highlight the necessity of ethical AI implementation, with a focus on transparency, human control, and support for critical thinking and creativity in educational settings.

Chapter 3: Learning Artificial Intelligence Through Open Educational Resources

Overview

Chapter 3, authored by Frédéric Alexandre, Marie-Helene Comte, Aurélie Lagarrigue, and Thierry Viéville, discusses the creation and impact of a Massive Open Online Course (MOOC) titled “Artificial Intelligence with Intelligence (IAI).” The chapter emphasizes the importance of fostering AI understanding among citizens and engaging them in ethical debates surrounding AI. It addresses three perspectives on integrating AI in education: adapting the learning experience, using AI as a scientific tool to study human learning, and critically examining AI’s problems.

Key Themes

1. Rationale Behind the MOOC:
   • The MOOC aims to cultivate a culture of AI understanding and participation in ethical discussions.
   • It highlights the importance of citizen training in AI, particularly concerning algorithmic decision-making ethics.
   • The course seeks to help people develop informed opinions about AI’s capabilities and limitations.
2. Three Perspectives on AI in Education:
   • Adaptive Learning Tool: AI can adapt learning experiences by analyzing student data and modifying difficulty levels or curating content. This approach enhances learner engagement and allows teachers to focus on more complex aspects of student learning.
   • Scientific Tool for Understanding Human Learning: AI can model human learning phenomena by analyzing learning analytics. This approach helps researchers understand how learning occurs and improves AI’s ability to support educational activities.
   • Critical Perspective: The chapter stresses the need for critical thinking about AI’s impact on education and society. It advocates for educating citizens about AI’s workings and its ethical implications.
3. Citizen Education and AI Literacy:
   • Citizen education is crucial for understanding AI technologies and their effects on daily life.
   • The chapter emphasizes the need for AI training that helps people comprehend how AI works and develop critical thinking skills.
   • The MOOC “Artificial Intelligence with Intelligence” serves as a tool to familiarize educators with AI and illustrate its role in developing skills.
4. Adaptive Learning in Practice:
   • The KidLearn Project is an example of adaptive learning, using algorithms to adjust learning activities based on the proximal zone of development.
   • Adaptive learning systems can enhance engagement and understanding by personalizing the learning experience.
   • However, there are concerns about data misuse, tracking, and the potential reduction of human staff in educational settings.
5. Challenges and Ethical Considerations:
   • The chapter highlights the potential risks of AI in education, such as the pervasive tracking and categorization of learners and exacerbation of inequalities.
   • It calls for designing AI tools that enhance human agency rather than restrict it.
   • Ethical AI implementation should prioritize transparency, human control, and support for critical thinking and creativity.

Summary

Chapter 3 emphasizes the importance of citizen education and critical thinking in the context of AI in education. The creation of the MOOC “Artificial Intelligence with Intelligence” aims to foster a culture of AI understanding and engagement in ethical debates. The chapter discusses three perspectives on AI in education: adapting the learning experience, using AI as a scientific tool, and critically examining its problems. It advocates for AI training that helps people understand how AI works and develop informed opinions about its potential and limitations. The chapter also highlights the potential risks and ethical considerations associated with AI in education, calling for tools that enhance human agency and support critical thinking.

Chapter 4: Digital Acculturation in the Era of Artificial Intelligence

Overview

Chapter 4, authored by Michel Durampart, Philippe Bonfils, and Margarida Romero, examines the integration of AI in education through the lens of digital acculturation. The chapter explores how the broad appropriation of digital tools impacts education, emphasizing the role of principals and the complex interplay between digital technology and education. It outlines the phases of digital acculturation, focusing on professional development and the motivations behind digital integration. The chapter also addresses the application of digital acculturation in various educational settings, including medical expertise, vocational training, and special education.

Key Themes

1. Definition and Scope of Digital Acculturation:
   • Acculturation is described as the process by which different cultural groups adapt to one another, encompassing the adoption of digital technologies.
   • The chapter highlights the importance of digital acculturation for educational stakeholders who must adapt to new educational practices involving AI.
2. Phases of Digital Acculturation:
   • First Level: Personal use of digital tools by teachers, focusing on knowledge acquisition through training. Many teachers use technology personally but face barriers to professional application due to insufficient training.
   • Second Level: Interaction between school and home, integrating digital applications into educational activities. This phase emphasizes the need for rehabilitation and instructional supervision to maximize the potential of digital tools in the classroom.
   • Third Level: Teaching using digital technologies, involving the use of digital tools in education. Teachers utilize various digital resources to enhance pedagogical practices and student engagement.
3. Challenges and Motivations:
   • The chapter discusses the emotional, psycho-cognitive, and pragmatic changes faced by teachers and students as they adapt to digital technologies.
   • It highlights the tension between innovative digital initiatives and traditional school organization, emphasizing the role of principals in facilitating necessary changes.
4. Applications in Different Educational Contexts:
   • Medical Expertise: AI is used to reorient medical proficiencies, saving time and facilitating expertise.
   • Vocational Training: AI and learning analytics are integrated into training programs to enhance individualization and self-learning practices.
   • Special Education: Digital technologies support learners’ performance, particularly in mathematics and reading, offering cognitive and behavioral remediation for students with specific needs.
5. Organizational Changes and AI’s Impact:
   • The chapter addresses the organizational changes required to incorporate AI technologies in educational platforms.
   • It emphasizes the need to reconsider how educators and learners adapt to and engage with data generated by AI tools.
   • AI’s integration into educational environments prompts new considerations about global education systems and the dynamics of learning and interfaces.

Summary

Chapter 4 delves into the concept of digital acculturation in the era of AI, exploring how the broad adoption of digital tools impacts education. It outlines three phases of digital acculturation, emphasizing the importance of professional development and the motivations behind digital integration. The chapter addresses the application of digital acculturation in various educational settings, including medical expertise, vocational training, and special education. It highlights the challenges and organizational changes required to incorporate AI technologies in educational platforms, emphasizing the need for educators and learners to adapt to and engage with AI-generated data. Overall, the chapter underscores the significance of digital acculturation in navigating the complex interplay between digital technology and education.

Synopsis of Chapter 5: Citizenship, Censorship, and Democracy in the Age of Artificial Intelligence

This chapter, authored by Tetiana Matusevych, Margarida Romero, and Oksana Strutynska, explores the ethical challenges posed by the integration of artificial intelligence (AI) in education and its implications for citizenship, censorship, and democracy. It underscores the importance of AI literacy, media literacy, and critical use of digital technologies to address issues such as information conflicts, political manipulation, and AI inequality.

The chapter discusses the need for comprehensive digital citizenship education to empower individuals to use digital tools responsibly and participate fully in society. Initiatives like the Digital Citizenship Education program by the Council of Europe aim to develop the values, skills, and knowledge necessary for responsible digital citizenship. The concept of AI for Social Good (AI4SG) is also highlighted, emphasizing the potential for AI to contribute positively to society, despite challenges in defining and replicating its benefits.

Several problems associated with AI integration in education are identified:

1. Polymorphic Nature of AI Ethics: AI ethics must consider both data and educational ethics, as the ethical use of AI extends beyond data management to encompass broader educational principles.
2. Threats to Fundamental Rights and Democracy: AI systems can perpetuate biases and oversimplify complex social realities, affecting human rights and democratic processes.
3. AI Colonialism: The dominance of AI technologies from specific cultural and commercial perspectives can marginalize diverse cultural contexts and reinforce inequalities.
4. Lack of Universal Ethical Regulation: Unlike healthcare, education lacks a universal approach to ethical AI use, leading to varied and sometimes insufficient ethical standards.
5. Ethics Washing: Companies may use superficial ethical guidelines to avoid meaningful regulation, undermining genuine ethical considerations.
6. Systematic Application of Ethical Principles: Ethical oversight in AI research and application is often inadequate, especially in educational settings.
7. Overuse of AI: Inappropriate or excessive application of AI can lead to dependency and cognitive decline, highlighting the need for balanced use.
8. Accountability and Responsibility: Determining accountability for AI-driven decisions in education is complex and often unclear.
9. AI Loyalty and Conflict of Interest: The interests served by AI systems must be transparent, with clear accountability to avoid conflicts of interest.
10. Decreased Social Connection: Increased AI use can reduce human interactions, exacerbating issues of loneliness and disconnection.
11. AI Threats to Citizenship: AI can be used for censorship and disinformation, manipulating public opinion, and exacerbating societal inequalities.
12. AI Misinformation: The potential for AI to generate and spread false information, including deepfakes, poses significant challenges to public trust and democracy.

The chapter concludes by emphasizing the dual nature of AI in education: while it poses significant risks to citizenship and democracy, it also offers opportunities to enhance education through adaptive learning, gamification, and support for creative and critical thinking. Ensuring that AI is used ethically and responsibly requires a deep understanding of AI fundamentals and a commitment to fostering digital literacy and critical engagement among educators and learners.

Summary of Chapter 6: International Initiatives and Regional Ecosystems for Supporting Artificial Intelligence Acculturation

Authors: Margarida Romero, Isabelle Galy, Jérémy Camponovo, Florence Tressols, and Alex Urmeneta

This chapter explores national and international initiatives to promote AI education and acculturation, focusing on various OECD countries’ efforts. It highlights the House of Artificial Intelligence (Maison de l’Intelligence Artificielle, MIA) in the French Riviera as a significant regional example. MIA supports AI education through partnerships, educational outreach, entrepreneurship initiatives, and tackling gender biases.

National initiatives vary, including Singapore’s adaptive learning system in mathematics and language education and India’s “AI for All” program, which aims to demystify AI for the general public. In the U.S., the Department of Education emphasizes spreading AI knowledge and involving educators in its policy report. At the university level, initiatives like France’s EFELIA support the development of AI fundamentals across various domains.

The MIA, an inclusive hub for AI education and innovation, offers activities for learners of all ages, teachers, and professionals. These activities range from using generative AI for educational purposes to showcasing AI technologies from local startups. The Arc-en-ciel project, part of MIA’s outreach, focuses on middle school students, addressing biases and gender stereotypes in AI. The project includes school visits, internships, and scientific mediation activities.

The chapter discusses the outreach curriculum developed by MIA, which includes AI demonstrations, school visits, and teacher training. It also emphasizes the importance of entrepreneurship in AI education, encouraging students to become creators rather than mere consumers of AI technology.

Addressing gender biases, the chapter highlights the Arc-en-ciel project’s efforts to promote gender equity in AI education. This includes creating a digital handbook to identify stereotypes, conducting studies on gender bias, and organizing intervention classes on International Women’s Rights Day.

The chapter concludes with the Smart Hive project, an interdisciplinary initiative aimed at fostering sustainable development through AI. MIA’s collaborative efforts with stakeholders in scientific, technical, and industrial sectors are crucial in promoting AI acculturation and ensuring inclusive and enriching educational experiences.

Summary of Chapter 7: Informal Education Practices for Human–AI Creative Pedagogy for Accessibility and Inclusivity

This chapter explores the educational initiatives of the Scientotheque, a Brussels-based association dedicated to promoting accessibility and inclusivity in technology for marginalized populations. Since 2020, the Scientotheque has focused on integrating artificial intelligence (AI) into education to address equity challenges. Key efforts include teacher training and AI workshops for disadvantaged youth, emphasizing the development of STEM skills and combating gender bias.

The Scientotheque’s mission is to support young people aged 4–20 through multidisciplinary STEAM (Science, Technology, Engineering, Arts, and Mathematics) workshops. These initiatives aim to reduce social inequalities and encourage interest in scientific and technological fields, particularly among girls. The COVID-19 pandemic highlighted the digital divide, prompting the Scientotheque to prioritize digital activities in its programs.

Collaborative partnerships are crucial to the Scientotheque’s efforts. The organization works with various schools and educational projects across Belgium and Europe, such as the Computer Learning Community (CAI), ESERO Belgium, and Dexterlab. These partnerships facilitate the development of innovative educational resources and teacher support systems.

The Scientotheque employs a project-based pedagogy for AI education, emphasizing hands-on experimentation, collaboration, and ethical reflection. In 2020, the association created a catalog of over 200 educational resources on AI, available under a Creative Commons license. These resources include activities on AI history, biological intelligence, programming, and ethical debates, designed for flexible integration into educational programs.

Teacher support is a critical component of the Scientotheque’s approach. The association offers ongoing assistance through hotlines, resource co-creation spaces, conferences, and networking opportunities. This support helps teachers implement new pedagogical strategies and fosters a collaborative learning environment.

The chapter highlights specific activities conducted during the 2020–2021 school year with young people from disadvantaged backgrounds. These workshops covered topics such as machine history, biological intelligence, algorithms, and AI’s societal impact. The hands-on, collaborative nature of these activities helped demystify complex AI concepts and encouraged critical thinking.

The Scientotheque’s initiatives address the growing need for AI education and aim to prevent deepening existing inequalities. The association’s efforts have been recognized at the federal level, leading to the AI4InclusiveEducation project. This project aims to develop and disseminate inclusive AI education content in French and Dutch, furthering the Scientotheque’s mission of promoting accessibility and inclusivity in AI education.

Chapter 8: Students’ Perspective on the Use of Artificial Intelligence in Education

Abstract: This chapter explores the perspectives of middle school students at Life Bloom Academy regarding artificial intelligence (AI) in education. The study focuses on students’ views before and after a semester-long AI acculturation program. The program included activities and a visit to the Maison de l’Intelligence Artificielle (MIA), where students engaged in multidisciplinary activities on AI applications. The study finds that students are aware of AI’s societal implications and maintain enthusiasm for the technology beyond the classroom.

Keywords: AI education, Students’ voice, AI risks, Citizenship, Agency, Early education

Introduction: Integrating AI in education requires considering students’ perspectives to ensure inclusivity and counter potential biases. This chapter reviews studies on higher education (HE) students’ views on AI and presents a study on middle school students’ perspectives after participating in AI acculturation activities.

Higher Education Students’ Perspective: HE students use AI tools like ChatGPT for structuring ideas, raising ethical and developmental concerns. Studies reveal a significant adoption of AI tools, particularly in STEM fields, with students appreciating AI’s assistance but wary of its potential to reduce human interaction and skill development. Concerns include the impact on interpersonal relationships, potential addiction, and over-reliance on technology.

Middle Schoolers Perspectives on AI: Middle school students also use AI tools, reflecting on AI’s potential to replicate human creativity and its ethical implications. The study at Life Bloom Academy involved preparatory sessions, visits to MIA, and interdisciplinary activities, focusing on developing critical thinking and understanding AI’s ethical dimensions.

Procedure: Students debated the nature of intelligence, participated in activities at MIA, and engaged in immersive AI activities back at the academy. They reflected on AI’s impact on daily life, privacy, social control, sustainable development, healthcare, and education.

Students’ Perception of AI: Students recognized AI’s ability to automate tasks and the importance of human involvement in training AI models. They expressed concerns about privacy, security, and potential misuse of AI, highlighting risks to personal security and individual freedom.

Students’ Concerns About Privacy and Social Control: Students emphasized the ethical implications of AI’s widespread data collection and the potential for misuse, stressing the need for robust safeguards and ethical considerations.

Students’ Perception of AI in Sustainable Development: Some students viewed AI as a solution to global issues, such as optimizing agricultural practices and managing natural resources, while emphasizing the need for ethical and sustainable implementation.

Students’ Perception of AI in Healthcare: Students had high expectations for AI in healthcare, particularly in care and treatment, but also raised concerns about data privacy and the risks of over-reliance on technology.

Students’ Expectations of AI in Education: Students identified potential uses of AI in supporting administrative tasks and personalizing the learning experience through real-time scaffolding and feedback.

Discussion: The study highlights the importance of integrating students’ perspectives in AI education. Both HE and middle school students recognize AI’s potential but are cautious about its ethical implications. The interdisciplinary project at Life Bloom Academy showed that students can critically analyze AI’s potential and risks, emphasizing the need for AI acculturation to develop their citizenship and agency in the age of AI.

Summary of Chapter 9: Affordances for AI-Enhanced Digital Game-Based Learning

Authors: Margarida Romero, Petros Lameras, Sylvester Arnab

Abstract: This chapter explores the integration of AI in Digital Game-Based Learning (DGBL), highlighting the interplay between pedagogy, social dynamics, and technological advancements. The authors examine how AI enhances educational experiences through personalised and adaptive learning. Key aspects include the influence of learning analytics, machine learning, and the practical impact of real-time feedback on DGBL.

Keywords: Learning analytics, Digital Game-Based Learning, Affordances, Feedback, Learner modeling

Introduction: The chapter introduces the concept of playful learning environments, specifically DGBL, enhanced by AI. The authors discuss the potential of AI to improve pedagogical and social affordances in these environments, focusing on the technical, pedagogical, and social aspects.

Design Affordances for AI Tools in Education: Game design in educational contexts requires alignment between game objectives, learning mechanics, and the playful learning experience. Affordances in game design can be categorized into three types: pedagogical, social, and technical. Effective game design supports the integration of these affordances, creating dynamic and adaptive learning environments.

Pedagogical Affordances in AI Tools for Education: AI in DGBL offers personalised and adaptive learning experiences. AI’s ability to adapt in real-time to learner profiles enhances data accuracy and provides actionable feedback. Pedagogical affordances include various approaches like problem-based, project-based, and game-based learning. AI enhances these approaches by offering customised learning experiences tailored to individual needs.

Social Affordances in AI for Education: AI can develop and leverage learner profiles to deliver tailored learning experiences. Social affordances focus on creating a conducive social context within ICT tools, ensuring safety, comfort, and convenience. AI can enhance social interactions by analysing emotions and ensuring a positive and inclusive online atmosphere.

Technical Affordances in AI for Education: Technical affordances pertain to the capacity and usability of ICT tools. Effective learning outcomes rely on the synergy between technical, pedagogical, and social affordances. AI enhances technical affordances by providing flexible and user-friendly tools that support personalised learning.

Affordances Perception, Learning Analytics, and Machine Learning: Learning analytics and machine learning play a crucial role in adaptive DGBL environments. Machine learning algorithms can enhance the analysis of learning analytics, providing insights to support adaptive learning. In digital game-based maths learning environments, machine learning has been used to identify learner behaviors and improve performance.

Discussion: The chapter highlights the importance of integrating pedagogical, social, and technical affordances, augmented by AI, to create effective and personalised learning experiences. Learning analytics and real-time feedback are crucial in adaptive DGBL, providing dynamic and continuous support to learners. The authors emphasize the need for a balanced approach to leveraging AI in education, promoting personalised and engaging learning environments.

Conclusion: The integration of AI in DGBL offers significant potential to enhance educational experiences. By leveraging learning analytics and machine learning, educators can create adaptive and personalised learning environments that cater to individual learner needs. The chapter provides valuable insights for educators, instructional designers, and developers in optimising digital learning environments.

References: The chapter cites various studies and examples, including works by Arnab (2020), Banihashem et al. (2023), Conati & Kardan (2013), and others, to support the discussion on AI-enhanced DGBL.

Summary of Chapter 10: Generative Artificial Intelligence in Higher Education

Authors: Margarida Romero, Jonathan Reyes, and Panos Kostakos

Overview

This chapter examines the impact of Generative Artificial Intelligence (GAI) on higher education (HE), particularly focusing on text and image generation tools like ChatGPT. It explores the applications of GAI in language learning, chatbot functionalities, and responsible AI implementation within the academic setting. The chapter highlights both the benefits and limitations of GAI, emphasizing the need for ethical deployment and comprehensive policy-making.

Introduction

The rise of GAI tools, notably ChatGPT, has significantly influenced academic research and policy-making in HE. These tools provide new opportunities but also present challenges, especially regarding policy adjustments and academic integrity. GAI models process vast datasets to generate human-like text and images, raising concerns about data privacy and potential misuse.

Uses of AI in Higher Education

The literature review by Baidoo-Anu and Ansah (2023) identifies several benefits of GAI in education, such as personalized tutoring, automated grading, and interactive learning. However, limitations include lack of human interaction, potential biases, and privacy concerns. Exploratory studies have demonstrated GAI’s utility in creating educational resources and providing personalized learning experiences.

Key Findings:

• GAI tools can assist in generating teaching units, quizzes, and personalized feedback.
• Potential disadvantages include reliability issues and lack of human touch in feedback.
• Studies highlight the importance of prompt engineering to maximize the benefits of GAI.

AI for Language Learning and Translation

AI translation tools, such as Grammarly and Quilbot, play a crucial role in enhancing accessibility and inclusivity in HE by overcoming language barriers. These tools are particularly beneficial in international academic environments, facilitating real-time translation and improving the quality of written work.

Chatbots in Higher Education

Chatbots, evolving from early models like Eliza, now incorporate advanced natural language processing (NLP) to support various educational tasks, including teaching, research, and administration. Modern chatbots can personalize learning experiences and provide immediate assistance, enhancing student engagement and support.

Implementation Examples:

• Guided learning paths
• Personal tutoring
• Study buddies
• Collaboration tools

Responsible Use of Generative AI Tools in Academia

The chapter stresses the importance of ethical guidelines for the use of GAI in education. The Russell Group universities provide a framework for responsible AI use, focusing on AI literacy, curriculum adaptation, and equitable access. Ensuring academic integrity and addressing ethical dilemmas are crucial for the responsible deployment of GAI.

Key Components of Responsible AI Use:

• Ethical guidelines and AI literacy programs
• Curriculum and assessment adaptation
• Policies for data privacy and accountability

Conclusion

The integration of GAI in HE presents numerous opportunities for personalized learning and administrative efficiency. However, it also necessitates careful consideration of ethical issues and robust policy frameworks to ensure responsible and effective use. The chapter highlights the need for ongoing evaluation and adaptation of AI tools to meet educational goals while maintaining academic integrity and inclusivity.

References

• Baidoo-Anu, D., & Ansah, L. O. (2023). Education in the era of generative artificial intelligence (AI): Understanding the potential benefits of ChatGPT in promoting teaching and learning. Journal of AI, 7(1), 52–62.
• Chan, C. K. Y. (2023). A comprehensive AI policy education framework for university teaching and learning. International Journal of Educational Technology in Higher Education, 20(1), 38.
• Rudolph, J., Tan, S., & Tan, S. (2023). ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? Journal of Applied Learning and Teaching, 6(1).
• European Commission. (2019). Ethics guidelines for trustworthy AI.

This summary encapsulates the chapter’s exploration of GAI’s transformative potential in HE, while addressing the ethical and practical challenges it presents.

Chapter 11: Artificial Intelligence in Professional and Vocational Training

Authors: Solange Ciavaldini-Cartaut, Jean-François Métral, Paul Olry, Dominique Guidoni-Stoltz, and Charles-Antoine Gagneur

Abstract

This chapter explores the integration of AI in professional training, highlighting the challenges related to learning analytics generated during activities involving humans or living entities. It addresses the intersection of initial and ongoing training and the design challenge of making AI tools acceptable for both training and the workplace. Through three case studies, the chapter examines the nature and quality of initial data and learning analytics using educational data mining. The studies focus on Comté cheese manufacturing, the Silva Numerica digital forest simulator, and automotive mechanics training. The chapter concludes by discussing data reliability concerns in AI systems and proposing strategies for education and training to address these challenges.

Keywords

Professional training, Vocational training, Learning analytics, Artificial intelligence, Digital learning environment

Introduction

Developing an AI system for professional training requires reliable data that supports teaching and learning processes. Vocational training emphasizes practical knowledge and adaptable workplace regulations. A key challenge is maintaining the educational value of real-life experiences while ensuring their applicability for training purposes. AI-powered smart tutors face obstacles in providing extended feedback for human interactions and accurately describing actual work processes.

The Use of Professional Analytics in Designing AI Tools

Professional skills and tasks are complex to document, especially within companies where proprietary work processes are involved. The challenge is to identify relevant data points for evaluating learning in professional environments, as actual work often involves a mix of discretionary and prescribed tasks. This section highlights the difficulties in obtaining tangible learning analytics and data faithful to work situations.

Case Study 1: Comté Cheese Manufacturing

Comté cheese production is well-documented, but understanding the essential variables influencing a cheesemaker’s decision-making for distinct flavors remains elusive. The inclination towards industrialized processes over contextual reasoning poses a socio-technical problem. The case highlights the challenge of modeling complex, dynamic environments due to inadequate information sensors and the prioritization of procedures over reasoning.

Case Study 2: Silva Numerica Project

The Silva Numerica project aims to create a digital learning environment for forest management. Despite efforts to document work processes, the complexity and contextual dependencies pose challenges. The project’s forest simulation introduces simplified representations to aid learning but faces limitations in achieving realism. The study underscores the need for AI to contribute to modeling training processes when dealing with poorly documented phenomena.

Case Study 3: AI in Automotive Mechanics Training

AI’s role in automotive mechanics training emphasizes the need for visibility in cognitive inference processes. The chapter discusses the challenges of providing meaningful feedback for learning in professional domains with diverse and context-dependent outcomes. AI’s feedback must account for various legitimate reasoning paths and not just align with standard practices.

Challenges and Potential of AI in Vocational Training

The chapter explores AI’s ability to generate pertinent feedback for learning and the interpretative challenges it faces. AI’s effectiveness in professional training depends on its capacity to detect pedagogically relevant discrepancies and tailor guidance accordingly. The limitations of relying solely on AI for learning maieutics and guidance based on discrepancies are highlighted.

Conclusion

The chapter concludes by emphasizing the need for reliable and tangible work data to unlock AI’s potential in vocational training. The evolution of AI presents challenges in tethering processes to reality, necessitating collaboration with professional organizations. Designing learning support tools that leverage work data through collaborative research is crucial. Effective AI integration in professional learning requires a multifaceted strategy and collaboration with educators and trainers to foster acceptance and usability.

References

The chapter provides references to various studies and works related to AI in professional and vocational training, including research on learning analytics, educational data mining, and case studies on Comté cheese manufacturing and forest management simulations.

This chapter presents a comprehensive exploration of the integration of AI in professional training, addressing the challenges and potential benefits of AI-powered tools in enhancing vocational education.

Chapter 12: Manifesto in Defence of Human-Centred Education in the Age of Artificial Intelligence

Authors: Margarida Romero, Thomas Frosig, Amanda M. L. Taylor-Beswick, Jari Laru, Bastienne Bernasco, Alex Urmeneta, Oksana Strutynska, and Marc-André Girard

Abstract
This chapter advocates for a balanced integration of AI in education, emphasizing a human-centred approach. It explores the potential and challenges of large language models (LLMs) and generative AI in transforming education. The authors highlight the importance of maintaining human-centricity in AI use, promoting ethical considerations, and empowering teachers and students as decision-makers. The manifesto underscores the need for responsible AI practices, regulations, and the concept of hybrid intelligence (HI) to enhance educational experiences.

Keywords
Human-centred education, Hybrid intelligence, Generative AI, AI education, Large language models

Introduction
The chapter opens with a historical perspective on AI, highlighting its shift from simulating human intelligence to disrupting human processes and practices. It discusses the rapid advancement of AI technologies and their impact on education, stressing the importance of balancing technological integration with human-centric educational values.

Empowering Students and Teachers as Decision-Makers

The authors emphasize involving teachers in the decision-making process of integrating technology in education. They argue that teachers play a crucial role in shaping the educational journey of students, and their professional expertise should be considered when implementing AI technologies. The focus is on giving educators autonomy and using data to support rather than overshadow pedagogic principles.

Impact of Artificial Intelligence on Existing Educational Paradigms

AI poses challenges to traditional educational practices and assessment methodologies. The chapter discusses the evolving role of AI as a partner in education, supporting self-regulation, collaboration, and problem-solving. It highlights the need for metacognitive skills and critical thinking when working with AI and stresses the importance of human oversight in AI-driven learning environments.

Artificial Intelligence in Human-Centred Education

The chapter advocates for ethical and responsible use of AI in education. It highlights the need for frameworks guiding AI integration, ensuring alignment between technological means and educational objectives. The European Union’s AI Act is presented as an example of such a framework, emphasizing human oversight and ethical considerations.

Hybrid Intelligence

Hybrid intelligence combines human and machine intelligence to enhance human capabilities. The chapter identifies challenges in collaborative, adaptive, responsible, and explainable HI. It argues for participatory approaches in developing data and AI literacy, fostering active engagement in human-AI collaborations.

Creative Uses of Artificial Intelligence in Education

AI can transform education by co-creating knowledge and reshaping human practices. The #PPAI6 model is introduced, presenting six levels of creative engagement in human-AI activities. The chapter emphasizes the importance of fostering agency and creative engagement through participatory and expansive learning supported by AI.

Inclusivity and Diversity in Artificial Intelligence

The chapter discusses the ethical and inclusivity challenges of AI in education. It calls for participatory design approaches and emphasizes the need for interventions that consider diverse educational actors. Teachers should be supported in their acculturation to AI, enabling informed decision-making on integrating AI technologies.

Advancing Towards an Increased Human-Centred Education in the Age of AI

The chapter concludes with a call for developing critical thinking and AI literacy among educational stakeholders. It highlights the importance of frameworks like the DigComp 2.2 for fostering digital competence. The manifesto emphasizes empowering teachers, ethically regulating AI use, and upholding the rights of educators and learners to create a balanced educational future.