Teaching with AI?

At the present moment, teaching with AI is a controversial issue, still. Some countries and organizational institutions have declared against the use of AI in education. I find this unthinkable. For me, it is like trying to stop the use of the Internet or any other technology that was, once, novel but time proved its usefulness from many perspectives – despite that some might use it unethically, at least in some contexts.

It is quite amusing to see these “controversies” lingering since the Council of Europe issued a quite useful document entitled ARTIFICIAL INTELLIGENCE AND EDUCATION. A critical view through the lens of human rights, democracy and the rule of law, distributed in November 2022.  This document aims “to provide a holistic view to help ensure that AI empowers and not overpowers educators and learners, and that future developments and practices are genuinely for the common good.” This report delves into the intricate intersections between AI and education, scrutinizing these dynamics through the prism of the Council of Europe’s commitment to safeguarding human rights, bolstering democracy, and upholding the rule of law. While recognizing the potential of AI in education, the exploration revealed a landscape rife with hyperbole, substantial challenges, and a plethora of questions.  The report challenges whether AI is designed to complement or replace teachers and if it personalizes learning paths to predefined content or supports personalized outcomes that cater to individual student aims and potentials. It constitutes a consistent shift of the interest in AI from technical aspects to a human dimension. It emphasizes the necessity to comprehend the diverse connections between AI and education, and to consider robust regulations.

This is only one of the reports issued in the European Union on AI (a list is provided in the annexes of the document). Also, UNESCO has proved early interest in AI and its potential in relation to education. It offers an online platform, entitled Artificial Intelligence and the Futures of Learning, which includes discussions of policies and various initiatives, as well as useful guides such as Guidance for generative AI in education and researchIt is important to mention that this platform is updated regularly. Therefore, it is a very useful database and forum for discussions. As in the case of European organisms, UNESCO also stresses the potential of AI, its usefulness, the need to stress the human dimensions, and the urgent need for regulation.

Considering the regulation, some steps have been initiated, such as the Organization for Economic Co-operation and Development (OECD) Principles on AI, which include recommendations for transparency, accountability, and the responsible use of AI. Some autoregulations have also been registered, as in the case of some technology companies that provide AI solutions for education, which have established their own ethical guidelines and policies for the use of their products in educational settings. Still, much progress needs to be made further, especially considering the changing landscape of AI.

We were curious to see what AI “thinks” about this aspect… Therefore, we asked ChatGPT to identify and explain the principles and recommended approaches found in documents related to AI in education. Here is the answer received (which actually reflects the most common aspects considered by the existing materials in the field):

ChatGPT: Certainly, let’s delve deeper into the common principles and recommended approaches found in documents related to AI in education:

  1. Ethical Considerations:
    • Principle: Ethical considerations are foundational to the use of AI in education. This involves ensuring that AI systems are designed, developed, and deployed with a commitment to fairness, accountability, and transparency.
    • Approach: Ethical AI frameworks often recommend conducting regular ethical impact assessments, involving stakeholders in the decision-making process, and addressing biases in algorithms to minimize unintended consequences.
  2. Data Protection and Privacy:
    • Principle: Protecting personal data and privacy is a fundamental concern. Users should have control over their data, and its collection and use should comply with relevant data protection laws.
    • Approach: Implementing privacy by design principles, obtaining informed consent for data processing, and adopting secure data handling practices are common approaches. Compliance with data protection regulations such as GDPR is often emphasized.
  3. Transparency and Accountability:
    • Principle: Transparency in AI systems is crucial to building trust. Users and stakeholders should have a clear understanding of how AI systems make decisions. Accountability mechanisms are essential to address issues or errors.
    • Approach: Providing explanations for AI decisions, disclosing the use of AI in educational settings, and establishing mechanisms for redress in case of errors are common approaches to enhance transparency and accountability.
  4. Inclusiveness and Equity:
    • Principle: AI in education should promote inclusiveness and equity, ensuring that all learners have equal access to opportunities and resources.
    • Approach: Implementing measures to address bias in algorithms, considering diverse perspectives in the development process, and designing AI systems that accommodate different learning styles and abilities are recommended approaches to support inclusiveness and equity.
  5. Human Rights and Non-Discrimination:
    • Principle: AI in education should respect and protect human rights, and it should not contribute to discrimination or exacerbate existing inequalities.
    • Approach: Conducting human rights impact assessments, actively preventing bias in AI algorithms, and addressing potential discriminatory effects in the design and deployment of AI systems are common approaches.
  6. International Collaboration:
    • Principle: Given the global nature of AI technologies, international collaboration is essential to develop common principles and standards.
    • Approach: Encouraging cross-border collaboration, sharing best practices, and participating in international forums and organizations to establish common guidelines contribute to a collaborative and globally aware approach.

These principles and approaches are not exhaustive, and specific recommendations may vary depending on the document, organization, or jurisdiction. However, they provide a foundation for ethical, responsible, and inclusive development and deployment of AI in education. As the field continues to evolve, ongoing discussions and refinements to these principles are likely to occur.

The question is not if we should use AI for educational purposes but how to use it. The answer is complicated by the dynamics of the AI environment, as well as the multitude of tools already existing and multiplying really fast. Also, there are so many different aspects to be considered. Here are only some of them: Neuroscience and AI, Educational content generation, Cognitive models and personalized learning, Adaptive and collaborative learning, Learning analytics, Automated assessment,  Interdisciplinary insights, Practical implementation of AI in educational contexts, Pedagogical innovations, Policies in Education, etc.

One of the good news related to using AI in education is that there is available support. For instance, ChatGPT – one of the most popular AI at the present moment – offers guidance, including suggested prompts, an explanation of how ChatGPT works and its limitations, the efficacy of AI detectors, and bias. Accessing this online guide, one can see examples and suggestions for effectively helping students to develop debates and role-playing, develop lesson plans and quizzes, or just to contribute to the development of critical thinking among students.  Essential to all these is to know, as a teacher, how to prompt ChatGPT and to teach students to prompt it effectively. Further useful aspects are offered through a series of Q&A.

Written by