Shape of the Future: Shape of the Future: How education system leaders can respond to the provocations of AI

I used Google’s Notebook LM to summarise this report. I have done this to bring the report and its main points to your attention, and to put Notebook LM through its paces.

Summary

This document provides a comprehensive review of the use of artificial intelligence (AI) in education, drawing insights from 23 Multi Academy Trusts (MATs) across England. The report examines the opportunities and challenges presented by AI, particularly in the context of generative AI tools like ChatGPT. It presents a MAT AI Guidance Framework with 10 key question sets designed to help educational leaders navigate the complexities of integrating AI into their schools. The document also highlights practical actions for system leaders to take, focusing on policy development, staff development, curriculum adaptation, safeguarding, and ethical considerations. The document aims to provide a roadmap for education system leaders to develop comprehensive strategies for AI integration, ensuring that students are prepared for success in an increasingly AI-driven world while maintaining the core values of education

Table of Contents: AI in Education - Insights and Recommendations for MAT Leaders

Executive Summary

This report summarises a project involving 23 Multi-Academy Trusts (MATs) exploring the integration of Artificial Intelligence (AI) in education. It introduces a MAT AI Guidance Framework, consisting of 10 key question sets, to guide strategic planning and implementation of AI. Key recommendations emphasise developing AI strategies, promoting AI literacy, adapting curricula, ensuring equity and access, addressing ethical concerns, and fostering research and collaboration. The report stresses the importance of a balanced approach that leverages AI to enhance, not replace, the human element of education.

Project Team (List of names representing various educational institutions)

Background

This section sets the context for the report, highlighting its purpose as a guide for school leaders navigating AI in education. It emphasises the need for safe, purposeful AI use by students and staff and clarifies the project's broad definition of AI, encompassing more than just generative AI tools like ChatGPT.

What do we mean by AI in schools?

This section elaborates on the definition of AI used in the project, drawing from the European Commission's definition. It stresses the importance of recognising the wide range of AI applications in education beyond generative AI and acknowledges the pervasive influence of AI in everyday life.

Research about AI in Education Pre ChatGPT

This section reviews research on AI in education predating ChatGPT. It highlights the potential benefits of AI in areas such as personalised learning, teacher workload reduction, and assessment. However, it also cautions about potential job displacement, ethical concerns, and biases, emphasizing the importance of context-specific implementation and rigorous evaluation.

Research about AI in Education after the release of ChatGPT

This section reviews emerging research on the impact of generative AI technologies like ChatGPT on education. While acknowledging the early stage of research, it identifies potential benefits for personalised learning, enhanced teaching practices, and improved efficiency. However, it also raises significant concerns around ethical implications, accuracy, potential skill deterioration in learners, equity, access, and teacher readiness, underscoring the need for balanced implementation, ethical guidelines, and ongoing evaluation.

The MAT AI Guidance Framework

This section introduces a 10-question framework designed to guide MATs in navigating the complexities of AI integration. The framework addresses strategic considerations, AI's role, governance, safeguarding, finance, technology, community impact, staff development, curriculum, assessment, equity, access, ethical implications, and monitoring.

Insights from Multi-Academy Trusts

This section presents findings from the 23 participating MATs, highlighting their diverse sizes, locations, and student populations. It emphasizes that while these MATs are at the forefront of AI implementation, they are still in the early stages of understanding and addressing AI-related issues.

Recommendations

This section presents 29 recommendations categorized under:

Considering how MAT leaders conceptualise AI

1. Tailor AI leadership resources to individual perspectives and priorities.

2. Encourage self-reflection on personal pedagogical and leadership beliefs.

3. Emphasise engagement with contemporary research.

Trends in relation to AI use by children and young people

1. Provide AI awareness training for students, families, and staff.

2. Conduct research on student use of generative AI outside school.

3. Research stakeholder perceptions of AI use in education.

Considering Knowledge, Accuracy & Reliability

1. Address AI bias and misinformation through education and training.

2. Examine the impact of technology on pedagogy, teacher satisfaction, and student learning.

Considering Safeguarding, Data & Privacy

1. Develop a comprehensive AI governance strategy for data protection and safeguarding.

2. Advocate for a centralised repository of Data Protection Impact Assessments (DPIAs).

3. Proactively address AI integration into existing educational technologies.

• Provides seven specific questions to guide risk mitigation

Considering Staffing & Workforce

1. Support research on AI's impact on the teacher's role and professional development.

2. Create a national working group to identify MAT functions that could benefit from AI.

Considering Curriculum, Assessment & Classroom Practice

1. Integrate AI literacy across all educational stages as part of curriculum review.

2. Adapt curriculum to prepare students for careers in an AI-driven job market.

3. Advocate for better categorisation of educational tools incorporating AI.

4. Conduct small-scale impact studies on AI use in schools.

5. Include key AI considerations in initial teacher education programmes.

Considering School Support

1. Utilize existing support networks for sharing best practices and challenges.

2. Diversify research approaches to include qualitative methods when evaluating AI.

3. Advocate for an independent body to filter and evaluate AI tools and resources.

4. Develop guidelines for using generative AI in school communications.

5. Engage with media to ensure nuanced and accurate reporting on AI in education.

Considering Policy (Local & National)

1. Support AI literacy and awareness training for education policy and accountability professionals.

2. Encourage policy leaders to define their roles and responsibilities regarding AI.

3. Revise the 1:1 device provision policy to address potential inequalities exacerbated by AI.

4. Initiate conversations about addressing inequalities in AI access between schools and MATs.

5. Advocate for improved quality assurance mechanisms for AI educational products.

6. Explore AI's potential in enhancing innovative schooling models and inter-model relationships.

Conclusion

This section summarises the key findings and emphasizes the need for a balanced, ethical, and strategic approach to AI integration. It underscores the importance of continuous evaluation, adaptation, collaboration, and research to shape a future where AI enhances human-centred education and prepares students for an AI-driven world.

References (Extensive list of academic references)

AI in Education: A Study Guide for System Leaders

Short Answer Questions:

1. How has the widespread use of the term "AI" since the launch of ChatGPT been potentially unhelpful in understanding its full range of applications in education?

2. What is the key advantage of AI-enabled adaptive learning environments, particularly for students' self-directed learning?

3. Describe two specific examples from mathematics education where AI-powered products have shown promise in improving student outcomes.

4. Despite the potential benefits of AI in language acquisition, what cautionary note did Sharadgah and Sa'di (2022) raise regarding their suitability for younger learners?

5. What concern does the 2020 A-level grading controversy during the COVID-19 pandemic highlight regarding the use of AI in educational contexts?

6. How can the concept of implementation research from the medical field be applied to understand the effectiveness of AI in education?

7. What are two key opportunities that generative AI technologies offer for enhancing the learning experience?

8. How can the implementation of AI-powered tutoring systems, such as NSWEduChat in Australia, address issues of educational inequality?

9. Despite the potential of AI to automate assessment tasks, what concern does Meniado (2023) raise regarding the accuracy of AI-generated responses?

10. What is a significant obstacle to the effective implementation of AI in education identified by Montenegro-Rueda et al. (2023) regarding teacher preparedness?

Answer Key:

1. Attributing "AI" solely to LLMs like ChatGPT overlooks other valuable AI applications and wrongly implies AI's presence is optional, ignoring its deep integration into our lives.

2. AI-enabled adaptive learning environments personalize the learning experience, allowing students to learn at their own pace, receiving tailored feedback and guidance with minimal teacher intervention.

3. Cognitive Tutor Algebra improved algebra proficiency over time, while ASSISTments, an AI-powered online math homework system, significantly boosted students' end-of-year math scores.

4. Many AI language tools aren't designed for younger learners and require careful integration into teaching, highlighting the need for age-appropriate implementation research.

5. The A-level grading controversy underscores how AI systems, if not carefully designed and implemented, can perpetuate or worsen existing inequalities in education.

6. Implementation research helps examine how the specific circumstances and actions surrounding AI technology implementation in diverse educational contexts impact its success or failure.

7. Generative AI can personalize learning through customized content and instant feedback and create more engaging learning experiences tailored to individual needs and styles.

8. AI-powered tutoring systems can provide equitable access to high-quality learning support, especially for students in underserved areas, bridging gaps in educational opportunities.

9. Meniado (2023) cautions about the potential for inaccurate responses from AI, emphasizing the need for human oversight and critical evaluation of AI-generated content in assessments.

10. A lack of adequate teacher training is a major hurdle, with many educators feeling unprepared to integrate AI tools effectively, highlighting the need for comprehensive professional development.

Essay Questions:

1. Critically analyse the ethical considerations surrounding the use of AI in education, particularly concerning data privacy, algorithmic bias, and the potential impact on student autonomy and teacher-student relationships.

2. Discuss how AI can be leveraged to address existing educational inequalities, ensuring equitable access to quality learning opportunities for all students, regardless of background or location.

3. Explore the potential impact of AI on the future of work and discuss how education systems can adapt curricula and pedagogical approaches to equip students with the skills needed to thrive in an AI-driven society.

4. Evaluate the role of school leaders in navigating the integration of AI in education, considering aspects such as vision setting, policy development, stakeholder engagement, infrastructure development, and ethical considerations.

5. Discuss the importance of ongoing research and evaluation in understanding the impact of AI on teaching and learning, proposing specific research questions that should be prioritized to guide future AI implementation in education.

Glossary of Key Terms:

AI (Artificial Intelligence): Systems exhibiting intelligent behavior by analyzing their environment and taking actions, with some autonomy, to achieve goals.

Generative AI: A type of AI capable of creating new content, such as text, images, or code, often based on large datasets and user prompts.

LLM (Large Language Model): A type of AI trained on massive text datasets to understand and generate human-like text.

AI-enabled adaptive learning environments: Digital learning platforms that use AI to personalize learning experiences based on student performance and preferences.

Data Protection Impact Assessment (DPIA): A systematic process to identify and minimize data protection risks when using new technologies or processes.

AI literacy: Understanding the capabilities, limitations, and ethical implications of AI technologies.

Algorithmic bias: Systematic errors in AI algorithms that can lead to unfair or discriminatory outcomes for certain groups of people.

Digital divide: The gap between those who have access to technology and those who do not, impacting access to educational opportunities and resources.

Implementation research: Research that focuses on understanding the factors that influence the successful implementation of interventions or innovations, such as AI in education.

FAQ: AI in Education

Strategy & Vision

1. How are Multi-Academy Trusts (MATs) in England approaching the integration of AI into their educational strategies?

2. While many MATs recognise the transformative potential of AI in education, they are still in the early stages of developing comprehensive strategies for its integration. Most leaders acknowledge that they haven't fully addressed all the opportunities and challenges AI presents. This is partly due to the rapidly evolving nature of AI technologies and the need for careful consideration of ethical implications, pedagogical approaches, and the human aspects of education. Current efforts primarily focus on raising awareness about AI among staff, students, and families, conducting small-scale pilot projects to evaluate the impact of specific AI tools, and developing governance frameworks to ensure responsible AI implementation. However, there is a clear need for more robust strategies that encompass curriculum adaptation, staff development, equitable access to AI technologies, and ongoing evaluation to inform future development.

3. What are some practical steps MAT leaders can take to address the strategic and ethical considerations of AI in education?

4. MAT leaders can adopt several practical actions to effectively integrate AI:

• Develop a comprehensive AI governance strategy. This should address children's rights, data protection, and safeguarding in the context of AI. Collaborate with children's rights organisations and legal experts to ensure compliance with relevant regulations.

• Advocate for a centralised repository of Data Protection Impact Assessments (DPIAs) specifically designed for educational technology tools. This will streamline compliance processes and ensure consistent data protection practices across schools and MATs.

• Provide AI awareness training for all stakeholders, including students, staff, and families. Focus on transparency regarding AI's capabilities and limitations, promoting ethical use, and addressing potential biases.

• Invest in staff development programmes to equip educators with the skills and knowledge needed to effectively integrate AI tools into their teaching practices.

• Review and update curriculum content to incorporate AI literacy and emerging skills. This will help students thrive in an increasingly AI-driven world.

• Conduct ongoing research and evaluation to assess the impact of AI on learning outcomes, teaching practices, and overall school operations.

AI Use & Impact

1. How are students currently using AI outside of school, and what are the implications for educational institutions?

2. While research on this topic is ongoing, anecdotal evidence suggests that students are increasingly using generative AI tools like ChatGPT for various purposes, including homework assistance, creative writing, and information gathering. This raises important considerations for schools regarding academic integrity, plagiarism, and the development of critical thinking skills. It also highlights the need for educational institutions to adapt to a landscape where AI is readily accessible and increasingly integrated into students' lives outside of formal education settings. Further research is crucial to understanding the motivations, patterns, and potential risks associated with students' independent AI use and inform educational strategies and policies.

3. How can schools ensure the accuracy and reliability of information and resources in an age of AI-generated content?

4. Addressing the proliferation of AI-generated content and its potential for bias and misinformation is crucial. Schools can mitigate these risks by:

• Incorporating AI and digital literacy into the curriculum. This will equip students with the skills to critically evaluate information sources, identify potential biases, and discern credible content from misinformation.

• Developing whole-community education programmes on AI, involving pupils, staff, and families. These programmes can raise awareness about the potential benefits and risks of AI, promoting responsible use and informed decision-making.

• Engaging with organisations developing quality assurance mechanisms for AI-related educational products. By participating in defining and evaluating "quality" in this context, schools can contribute to a more trustworthy and reliable AI marketplace for education.

Future of Education & Work

1. How can schools prepare students for a future workforce increasingly influenced by AI?

2. Equipping students with the skills and knowledge needed to thrive in an AI-driven world is paramount. Schools should consider:

• Adapting curricula to incorporate AI literacy, computational thinking, and problem-solving skills. These skills are transferable across various fields and crucial for navigating the complexities of an AI-augmented society.

• Cultivating skills likely to be in high demand, such as data analysis, machine learning, and AI ethics. This may involve partnering with industry experts or offering specialised courses in these areas.

• Fostering creativity, critical thinking, and collaboration skills, which remain essential in any work environment. AI should be seen as a tool that can augment, not replace, human capabilities.

• Providing career counselling and guidance that reflects the evolving job market and the rise of AI-related professions.

Policy & Support

1. What role can policymakers play in supporting the responsible and effective integration of AI in education?

2. Policymakers have a crucial role in shaping the AI-powered educational landscape by:

• Providing clear guidelines and frameworks for the ethical use of AI in educational settings. This includes addressing data privacy, algorithmic bias, and accountability for AI-driven decisions.

• Investing in research and development of AI technologies specifically designed for educational purposes. This includes funding for innovative AI tools, resources, and pedagogical approaches.

• Supporting teacher training and professional development programmes focused on AI literacy and the effective integration of AI tools in the classroom.

• Promoting equitable access to AI technologies and resources for all students, regardless of socioeconomic background or geographical location.

Resources & Collaboration

1. What resources and support systems are available to schools navigating the challenges and opportunities of AI in education?

2. Schools can leverage existing support networks and resources:

• National networks, such as EdTech Hubs and Challenge Partner Trust Leaders, offer access to shared experiences, best practices, and practical guidance from peers across the country.

• Collaborate with other schools and MATs to share resources, expertise, and lessons learned in AI implementation.

• Engage with organisations, such as BESA and the Teacher Development Trust, which provide resources, research, and professional development opportunities related to AI in education.

Evaluation & Monitoring

1. How can schools effectively evaluate the impact of AI on teaching and learning, and what should this evaluation encompass?

2. Evaluating AI's impact in education requires a nuanced approach that combines quantitative and qualitative data:

• Measure the impact of specific AI tools on student learning outcomes. This might involve comparing the performance of students using AI-powered learning platforms to those using traditional methods.

• Gather feedback from teachers on how AI tools are impacting their teaching practices, workload, and student engagement.

• Assess the impact of AI on student motivation, engagement, and self-directed learning.

• Evaluate the ethical implications of AI use, including potential biases, data privacy concerns, and the impact on student-teacher relationships.

Regularly review and adjust AI strategies based on the evidence gathered, ensuring that AI implementation aligns with the school's educational goals and values.

AI-generated discussion