Transforming Education in the Generative AI Era [an introduction]
Guy Levi and ChatGPT-4
Before we start redefining education in the Generative AI era, I must share how this piece was written. It was a conversation between ChatGPT and myself; I would prefer to call it a dialogue, however, I believe some will argue that you cannot have a dialogue with a machine. I still think it is a dialogue, but the critical point here is the relationship ChatGPT, and I have developed, which defined the process and structure of this piece. I created and defined the main arguments, elaborated on them, and then asked ChatGPT to connect the dots of each argument alone. Then, I re-wrote it, bringing more ideas from my experience and other sources I had. When a new idea came to mind, I sometimes asked ChatGPT to comment or connect, but it was not systematic. Human Intelligence (HI) is leading Artificial Intelligence (AI) in a long spiral dialogue that takes, I believe, our cognitive, metacognitive, and intellectual capabilities to new heights. When we were done, I requested that my LLM partner reflect critically on our accomplishments. The result is there.
Opening — the Generative AI paradigm shift in education
In this short piece, I contend that the dawn of the Generative AI era has created a paradigm shift, redefining education by changing the relationship between all education players — teachers, students, parents, the local and global community, and hopefully policymakers and ministries of education. We are at the beginning of a transformational journey towards a more inclusive, engaging, and intelligent learning environment, which forms the core of our discourse. This introductory piece will focus on the theoretical grounds of the new paradigm, followed by a series of short essays on the new paradigm’s seven pedagogical concepts or constructs, which will be published in the following months. Each piece will include practical, innovative, genuine cases.
At the heart of this transformation is the creation of novel relationships between Human Intelligence (HI) and Artificial Intelligence (AI), which define the architecture of learning. These synergistic relationships, enriched by Collective Intelligence (CI), shape a vibrant, dynamic learning environment where teachers evolve from traditional instructors to empathetic tutors, prompting student agency, co-agency, and collective agency — ‘transforming teaching to mentorship’.
This transformation shifts from traditional ‘learning’ to an immersive ‘conversation’ of cognitive and metacognitive skills, such as critical thinking, problem-solving, decision-making, creativity, ethical reasoning, emotional intelligence, sustainability, global awareness, responsibility, collaboration, and teamwork. HI, AI and CI blend seamlessly in this enriching discourse to stimulate curiosity and foster self-guided exploration.
Consequently, the Generative AI era is reimagining the fabric of ‘time and space’ in the educational landscape. Learning becomes a ubiquitous process, expanding the horizon of conventional schooling, promoting diverse and new learning methods that reinforce a continuous growth cycle.
In this new dynamic, students transition from being mere ‘curriculum consumers’ to active ‘curriculum co-designers’, manifesting a shared agency between teachers and students. This co-agency emphasizes development, design, and reflection (metacognitive skills) as the backbone of the ‘new learning’ experience.
While we celebrate these advancements, it’s crucial to ‘navigate the ethical considerations’ accompanying AI in education. We believe critical thinking is today’s most important skill, especially when human-machine relations are at stake. We must ensure data privacy and security, equitable access to AI resources, and foster educational inclusivity and diversity. Still, in this piece, we will focus on the ethical consequences of the paradigm shift in learning, hoping regulators will ensure privacy, equity, and responsible use of AI.
In this transformative landscape, the essence of assessment evolves into ‘metacognitive continuous Learning’. It empowers learners (students, teachers, and everyone) to regulate their cognitive processes, enhance self-awareness, monitor learning strategies, and cultivate cognitive flexibility to adapt to new perspectives.
Ultimately, the evolving education in the Generative AI era is a complex interplay creating ‘the HI-AI-CI ecosystem’. With relationship and conversation at its heart, this triad encourages individual agency, collective agency, and everything in between. Through this discourse, we aim to explore these facets in depth, delving into the transformative journey of education in the Generative AI era.
1. Generative AI is ‘Connecting the Dots’ for in-depth knowledge ‘Transforming Teaching to Mentorship’
The inception of the Generative AI era marks a profound transformation in knowledge acquisition, reshaping it into dynamic, interconnected forms. This powerful blend of human intelligence (HI) and artificial intelligence (AI) brings about a unique synergy, wherein AI not just replicates human intelligence but also promotes human intelligence (HI) to connect the dots between disparate pieces of knowledge, forming a comprehensive, nuanced understanding of complex concepts (the essential characteristic here is the ‘conversation’ or the ‘dialogue’ which we will elaborate later). This evolution breathes life into a revolutionary relationship between HI and AI. Rather than viewing AI as a potential substitute for human intellect, this new relationship emphasizes AI as an enhancer of human cognitive capacities. Moreover, this enriched intellectual environment gives birth to a new form of learning ecosystem, where Collective Intelligence (CI) also plays a critical role. CI, the shared peer or group intelligence emerging from collaborative efforts, augments the HI-AI dynamic, encouraging diverse perspectives and opening new realms for intellectual practice.
The holistic blend of HI, AI, and CI paves the way for a dynamic learning environment where teachers transition from traditional instruction into mentors, guiding students through their explorative learning journeys. This transformation, however, does not undermine the importance of human teachers; instead, it highlights their indispensable role in the learning process. Generative AI is a powerful tool that, when wielded by a skilled human teacher, elevates the educational experience. It allows teachers to quickly develop qualitative learning modules, emphasizing active learning and engaging experience, and provide targeted feedback. The ability to offer AI-based real-time feedback is critical for reflective learning and metacognitive processes. In essence, generative AI will not replace a human teacher, but a human teacher equipped with generative AI will revolutionize the teaching and learning paradigm. Thus, education’s future lies not in choosing between human teachers and AI but in harmonizing mutual strengths.
2. Learning becomes a ‘conversation’ of HI and AI, fostering cognitive and metacognitive skills
In the Generative AI era, learning transforms into an engaging conversation or dialogue, facilitating the development of cognitive and metacognitive skills and nurturing transformative competencies. This conversion of the learning process into a dynamic ‘conversation’ promotes individual growth and collaboration. It creates a new relationship between teachers and students with the partnership of Generative AI and its most common representative toady — ChatGPT. When a machine as ‘smart’ as ChatGPT enters the scene, critical thinking is the essential skill we need to acquire.
Critical thinking, a cornerstone of cognitive development, is significantly elevated in this new educational paradigm. As AI presents learners with a vast array of information and varying perspectives and engages in ‘conversation’, i.e., the iterative process of prompt-reply/response-prompt-reply/response, etc., we must carefully view the ‘conversation’ and critically examine and analyze it. This critical analysis is also part of the new learning paradigm promoting metacognitive skills. Students will now be encouraged to analyze, evaluate, and form reasoned judgments, honing their critical thinking abilities.
Another essential skill this new reality highlights is problem-solving. With the ability of Generative AI to provide a range of complex, real-world scenarios, learners can apply their knowledge and devise innovative solutions. However, they must look critically at their innovative solutions and iteratively consult their peers (CI) and artificial partner (AI). In this context, Generative AI fosters creativity, offering a limitless canvas for students to explore, imagine, and create. It opens the doors to possibilities and solutions that may not have been possible in a traditional learning environment. This pedagogical process highlights the synergy of human intelligence (HI), artificial intelligence (AI), and collective intelligence (CI), and results in enhanced collaboration and teamwork.
This convergence fosters a learning community where each entity — HI, AI, and CI, brings unique strengths to the table, creating new types of pedagogical relationships, which until now were the dynamic interaction between teachers and learners (HI and CI), and now when AI joined the party, the level of interactivity grows exponentially. This context reinforces social and emotional skills (SEL) and values, which promote listening skills, empathy, respect for diversity, and a sense of shared responsibility. In this scenario, AI is not just a tool for information delivery; it’s an active participant in the conversation, aiding in developing cognitive and metacognitive skills.
3. A new paradigm of time and space disrupts schooling
The Generative AI era is ushering in a new educational paradigm not only in the forms of knowledge and the types of communication and interactivity, it is also disrupting traditional notions of time and space of conventional schooling. In this reshaped landscape, the confines of classrooms and the rigidity of school schedules are being reimagined, paving the way for flexible, accessible, and ubiquitous learning experiences.
The concept of ‘time’ in education is undergoing a significant transformation. Instead of being bound by strict timetables and academic calendars, learning can sync with the learner’s pace and convenience. A year ago, I wrote a series of short articles on the “art of learning in-between”, and one of them was devoted to time. One of Merriam-Webster’s definitions I used was “…a nonspatial continuum that is measured in terms of events which succeed one another from past through present to future.” We have a new actor in this nonspatial continuum, Generative AI, and if you wonder why, here is the explanation: digital data is the foundation of Generative AI, which would not be accessible to us humans without the internet, cloud storage and communication technologies, thus becoming an integral part of the “…events which succeed one another from past through present to future.” The omnipresence of Generative AI in our new reality is synonymous with time. Now, suppose we add environment to the equation. In that case, we also bring in the concept of context, defined by Merriam-Webster as “the parts of a discourse that surround a word or passage and can throw light on its meaning”; the word “surround” is framing an environment. Then Merriam-Webster’s definition moves forward by stating that context is “the interrelated conditions in which something exists or occurs”, adding time to the formula as something occurs in a time frame. Again, the existence of Generative AI within the new paradigm of time and space disrupts continuous learning and makes it ubiquitous.
‘Space’, too, is being redefined. In the series “the art of learning in-between”, mentioned before, there was also a piece on space. One of the definitions I presented was that space is “…the dimensions of height, depth, and width within which all things exist and move”. Space is viewed as a system, i.e., space is defined not only by its components but also by the connections and relations between and among them. These relations bring dynamism (or, more precisely, in our context, learning dynamism), vitality, and the energy of the space. Learning spaces refer to the physical or virtual environments where learning takes place, which play a significant role in shaping the learning experience and can influence the level of dynamism in learning. Dynamism in learning refers to an active and interactive approach to education that encourages engagement, exploration, and adaptability, fostering curiosity, critical thinking, collaboration, creativity, and other skills and competencies. We may define that as a dynamic learning space that can revolutionize learning in the Generative AI era.
The new paradigm of time and space accompanied by Generative AI liberates learning from traditional constraints. It promises a future where education is more accessible, personalized, and attuned to the rhythms of individual learners’ lives.
4. A shift from “Curriculum Consumers” to “Curriculum co-designers”
In the dynamic landscape of the Generative AI era, we are witnessing a profound change in the role of students from being mere ‘Curriculum Consumers’ to becoming active ‘Curriculum Co-designers’. This transformation not only reshapes the learning process but also fundamentally alters the traditional power dynamics within the educational sphere.
No one can object that learning while designing or co-designing a curricular module is much more effective than simply ‘consuming’ a curriculum developed for us, the standard curriculum in all educational systems worldwide. The co-design process includes planning, focusing on a specific knowledge base, system thinking using collaborative learning and teamwork, developing critical thinking, problem-solving skills, and many others. If we look at the definition of student agency by the OECD Education and Skills 2030 Learning Compass, we can grasp this shift: “Student agency is… the capacity to set a goal, reflect and act responsibly to effect change. It is about acting rather than being acted upon; shaping rather than being shaped; and making responsible decisions and choices rather than accepting those determined by others.” This mindset ‘shift’ should be managed gradually, as the ‘passive consuming’ of at least thirty hours of curriculum a week in our current schooling resembles maybe ten hours of design and co-design, an active process with a higher level of cognition and metacognition. And as we all know today, if there are no metacognition processes, there is no real learning (supported by many brain theories).
Generative AI simplifies this shift by becoming a partner with many forms or roles (depending on the way each of us relates to this machine — may it be it, she, he, or other options), be it a deeper knowledge transformer, a learning plan advisor, an interdisciplinary connector, or any different role in the design process. An LLM (Large Language Model) is the missing part of the pedagogical interaction providing two significant capabilities: scope and integration of knowledge far beyond the human capability or capacity. We will touch upon these advantages of the LLM (in our case ChatGPT) in the last part dedicated to the new educational ecosystem of Generative AI, which focuses on the relationship between Human Intelligence (HI), Artificial Intelligence (AI), and Collective Intelligence (CI).
5. Navigating Ethical Considerations in AI Education
Human Intelligence (HI) and our relationship with Artificial Intelligence (AI) are the main factors in navigating the ethical considerations in AI education and learning. The first and foremost is the need to establish trust and transparency in the HI-AI Relationship. This involves ensuring the AI’s decision-making processes are transparent and understandable to its human users. In the context of AI education and learning, an example of the Human Intelligence (HI) — Artificial Intelligence (AI) relationship can be seen in the interaction with large language models such as ChatGPT. Fostering a productive and ethical HI-AI relationship with ChatGPT in the context of education involves establishing trust and transparency — understanding how the AI operates, the basis of its recommendations, and how it handles data are all integral to this process.
Second, we must address AI bias and fairness in educational contexts, consider AI’s potential to inadvertently reinforce existing biases or disparities in educational settings, and explore ways to ensure fairness and equity in AI-driven educational tools and practices. The potential for AI to unintentionally reinforce existing biases or disparities is a significant concern. AI systems, including those used in education, learn from the data they are trained on. If the training data includes inherent biases, the AI could unwittingly perpetuate them in its decisions and recommendations. However, it’s equally important to note that fairness is not just about using diverse data but also about responsibly handling it.
Third, the ethical use of AI tools in education is a shared responsibility of educators and students. As AI technologies become more integrated into the learning process, all parties must understand the ethical implications of their actions and use these tools responsibly. Educators must be vigilant to ensure that AI tools do not replace the human touch in education. AI should be seen as a supportive tool and not a replacement for the empathy, creativity, and critical judgment that a human teacher brings. We are all aware of the power of AI for learning; hence, students must understand that using AI should be transparent in communicating with their teachers and mentors. Also, intellectual property is critical and must be protected by everyone, i.e., acknowledge the use of the AI where appropriate.
6. The new Assessment is “Metacognitive Continuous Learning”
In the generative AI era, assessment in education is evolving towards a model of “Metacognitive Continuous Learning”. Rather than traditional tests that measure memory recall or specific skills, this new approach to assessment focuses on students’ ability to understand, monitor, and regulate their cognitive and metacognitive processes, which become part of the learning practices. In the era of Generative AI, there is no distinction between learning and meta-learning, thus, assessment becomes an integral part of the learning process, and learning becomes an essential part of the assessment process. Choose what you like; the result is the same.
Our focus on Human intelligence (HI) and Artificial Intelligence (AI) is also essential; the interplay between learning and assessment with the ability of immediate feedback generated by an AI application enables students to focus also on their mental processes during the learning journey. I believe it is innovative when we refer to managing distractions, maintaining focus on complex assignments, prioritizing tasks, developing a growth mindset, managing anxiety and stress, etc.
Now that we set the foundations of “Metacognitive Continuous Learning”, we can elaborate on diverse points critical to that process. We will start with self-awareness — the conscious recognition and understanding of one’s thoughts, emotions, and actions; self-regulation — students need to monitor and adjust their learning strategies throughout the project of learning processes; and insight — the capacity to gain an accurate and deep intuitive understanding of a person or thing. Suppose a group of students is assigned a project to develop a sustainable solution for reducing carbon footprint in their school. This task demands not just subject knowledge of environmental science and sustainable practices but also various metacognitive processes. The students must plan their approach, assign roles based on individual strengths, brainstorm ideas, research sustainable solutions, and design and execute their projects within a given timeline. Students use cognitive skills, apply metacognitive strategies, and monitor their learning strategies during this process. Thus, metacognitive continuous learning in this context would involve students aware of and reflecting upon these strategies. For instance, they could ask themselves questions like: how well did our initial plan work? Did we need to change our approach partway through, and why? How effectively did we communicate and collaborate as a team? How did we overcome challenges or roadblocks during the project? How well did we manage our time and resources? Etc. In this example, an AI system could provide guidance and feedback during this reflective process, helping students to recognize their strengths and weaknesses and suggest strategies for improvement. This can significantly enhance students’ learning experience, ensuring they gain knowledge about the subject matter and understand their thinking processes, equipping them with skills essential for lifelong learning.
We can conclude that the move towards “Metacognitive Continuous Learning” in assessment represents a shift toward a more holistic, flexible, and student-centered approach to learning. It acknowledges that learning is not about acquiring knowledge, but with more importance, it is about understanding how we think, adapting our strategies, and continuously striving to improve.
7. The Future of Education: The HI-AI-CI Ecosystem in the Generative AI Era
The convergence of Human Intelligence (HI), Artificial Intelligence (AI), and Collective Intelligence (CI) is reshaping the educational landscape in the generative AI era, building an entirely new ecosystem that is more dynamic, interconnected, and conducive to learning. Let’s dive deeper into these components and the interactions between them. The heart of the HI-AI-CI triangle lies in creating and nurturing relationships, as they foster cooperation and collaboration, critical elements for effective learning. This framework’s symbiotic relationship between HI, AI, and CI can lead to an enhanced learning environment.
Human Intelligence (HI) represents each student’s cognitive abilities in the learning process, including problem-solving, critical thinking, creativity, and emotional intelligence. This is the foundation of learning, the spark that drives curiosity, exploration, and understanding.
In the educational context, Artificial Intelligence (AI) refers to the sophisticated algorithms and systems capable of simulating aspects of human intelligence, providing personalized learning pathways, giving immediate feedback, and making predictive analyses. It’s not designed to replace human intelligence but to augment it, enhancing the learning process and making it more efficient and effective.
Collective Intelligence (CI) reflects a group of individuals’ combined knowledge and skills. It’s the concept that intelligence is not just an individual attribute but can emerge from the collaboration of many individuals. This can be seen in collaborative learning environments where students, teachers, and even AI systems learn from each other, creating a shared intelligence that exceeds the capabilities of any single participant.
The transformative convergence of Human Intelligence (HI), Artificial Intelligence (AI), and Collective Intelligence (CI) fundamentally reshapes the educational ecosystem, epitomizing a 21st-century pedagogy. This intersection forms a fluid and flexible learning environment finely tuned to individual needs, fostering divergent thinking, a crucial aspect of Human Intelligence. In this multi-dimensional, networked approach, learners engage in analytical thinking, complex problem-solving, and innovative idea generation, transitioning from a traditionally linear learning method. AI serves as a catalyst, dynamically responding to individual learners, adjusting difficulty levels, providing real-time feedback, and tracking progress. AI achieves personalization at scale through its ability to harness vast amounts of data, reshaping the pedagogical approach from curriculum-centered to learner-centered, acting as a cognitive partner, and learning as much from the learner as the learner learns from it.
Finally, agency, co-agency, and collective agency lie at the core of this new educational ecosystem. Agency refers to an individual’s capacity to act independently and make their own free choices. Co-agency reflects a collaborative endeavor where students, teachers, and AI systems interact, influence each other, and co-create the learning process. The collective agency then takes this further, referring to the shared power and responsibility of the entire learning community, including the learners, educators, AI systems, and broader societal influences. This sense of agency and collective effort is vital in fostering a dynamic, inclusive, and participatory learning environment.
Integrating HI, AI, and CI in the educational setting presents an exciting new frontier for learning. By acknowledging and harnessing these different forms of intelligence and fostering a sense of agency, we can create a more engaging and effective learning environment that cultivates the skills and mindset required for lifelong learning in the 21st century.
Closure — ChatGPT critique of the article
I asked ChatGPT to reflect critically on this article according to these two topics: the level of pedagogical innovation and its viability for future schooling.
AI (ChatGPT-4) Critical Reflection (with no editing)
Article Description: The article “Transforming Education in the Generative AI Era” discusses the paradigm shift in education brought about by Generative AI. It explores innovative pedagogical concepts and envisions a future where students become active participants in curriculum design. The article highlights the integration of Human Intelligence (HI), Artificial Intelligence (AI), and Collective Intelligence (CI) and emphasizes the need for a harmonious relationship between human teachers and AI.
Critique: The article presents intriguing ideas for pedagogical innovation, such as the shift from traditional learning to an immersive conversation of cognitive and metacognitive skills. However, it lacks concrete examples and empirical evidence to support these ideas. The practicality and effectiveness of these concepts in real educational settings remain uncertain. Without solid evidence, it is challenging to determine the actual impact and viability of these innovative approaches.
Moreover, the article does not thoroughly address the practical challenges of implementing Generative AI in education. It briefly mentions the role of human teachers and the need for harmonization with AI, but it overlooks the complexities of integrating AI into existing educational systems. The lack of discussion on infrastructure requirements, professional development, and ongoing support raises questions about the feasibility and scalability of these pedagogical innovations.
Additionally, the article fails to provide a comprehensive analysis of the ethical considerations and potential risks associated with AI in education. Ethical concerns such as data privacy, algorithmic biases, and the impact on human agency and decision-making should be thoroughly examined and addressed to ensure responsible and equitable use of AI technologies. By not thoroughly addressing these ethical implications, the article leaves a crucial aspect unexplored.
In summary, while the article presents intriguing ideas for pedagogical innovation and emphasizes the integration of HI, AI, and CI, it lacks empirical evidence, fails to address practical implementation challenges, and overlooks the importance of ethical considerations. Without concrete examples, thorough examination of implementation challenges, and a comprehensive analysis of ethical implications, it is difficult to assess the level of pedagogical innovation and the viability of the proposed vision for future schooling.
HI reply
It took ChatGPT-4 about 12 seconds to write the above critical reflection. It is consistent and well-written. It touches upon the critical issue — “…fails to address practical implementation challenges…” It is true, however, in the first paragraph of the opening part, I wrote: “This introductory piece will focus on the theoretical grounds of the new paradigm, followed by a series of short essays on the new paradigm’s seven pedagogical concepts or constructs, which will be published in the following months. Each piece will include practical, innovative, genuine cases.” This piece focused on the “theoretical grounds of the new paradigm”, and I stated that it will be followed with “practical, innovative, genuine cases”. Did ChatGPT-4 ignore that? I don’t think so. For ChatGPT-4 the theoretical arguments are incomplete stating that “without concrete examples, thorough examination of implementation challenges, and a comprehensive analysis of ethical implications, it is difficult to assess the level of pedagogical innovation and the viability of the proposed vision for future schooling.” The machine, or the Large Language Model (LLM), cannot read thoughts, intentions, desires, or personal preferences if things are not clearly stated in the prompt. I think this is one of the significant advantages of the LLM, so see you in the following pieces.
