Education In An Age Of A.I.
Updated: May 8
Education just shifted. Radically! Did you notice?
Let’s be honest. We have all long known that the educational system is out-dated and largely ineffective, but it is so hard to leave the system.
If you are parent, it seems you might be jeopardizing your child’s future and preparedness by going some experimental route.
If you are a teacher or a school, it means changing the entire approach. Out with old, in with the new. That is a tall order.
So we go along, even though the system has not been serving our children for quite some time now.
But this... is entirely different.
Artificial intelligence is not just one of those new tech catch-phrases that we may want to ignore. This is a paradigm shift of profound proportion, and it is happening at such a rate that career paths are literally becoming obsolete by the day. Where does that leave us? And more to the point, where does it leave our children and our educational system?
The capabilities that AI now brings to bear are able to take over many, if not most, of the pedagogical tasks that the old educational paradigm expected of us as students. It can already do pretty much any homework assignment, answer any essay prompt with appropriate stylistic analysis and references, and even generate our college essays, cover letters, and official resumes. Anything from within the existing knowledge base and anything iterative of typical human endeavors is a slam dunk for this 4-month-old technology. While the quality of each result may not yet be quite up to par, and while many teachers may still be able to tell the difference, the tools are only going to get better and more powerful. And far more quickly than we might think.
This goes far beyond issues like plagiarism or pedagogical styles. New plug-ins can now also connect these AI tools to powerful computational functionality and real-time internet data. The Wolfram plug-in, for example, gives ChatGPT the functionality to access “computation, math, curated knowledge and real-time data” from Wolfram Alpha and Wolfram Language. Constructing and then using a creative computational model — that we might not have even been able to think of — can now be done, behind the scenes, by AI when answering our question! The possibilities become truly endless.
There is no point in us trying to replicate or compete with AI. The only course of action is to leverage it.
As such, the issue of our educational system has just become an immediate and pressing one. It is no longer merely a question of efficacy but, in fact, a question of triage. If we do not immediately shift our educational focus, those currently being taught in the old paradigm risk graduating into a world that has already left their mindset and skill sets behind.
So, if we must make a change, what might this new approach to education look like in an age of AI? Instead of taking the time to compile a list of elements that might constitute such a new approach to education, I thought I’d first see what ChatGPT-4 had to say on the subject. (It’s a lot quicker!)
I posed the prompt: "What should the purpose and focus of a high school education be in the age of artificial intelligence?"
In the age of artificial intelligence (AI), the purpose and focus of a high school education should adapt to the changing landscape of technology, job markets, and societal needs. Here are some key areas that high school education should emphasize:
Digital Literacy and Computational Thinking: a. Digital literacy involves teaching students how to use digital tools, access and evaluate online information, navigate social media responsibly, and understand online privacy and security. Integrating these skills into the curriculum helps students become competent and responsible digital citizens. b. Computational thinking encourages students to approach problems systematically, using techniques like algorithmic thinking, pattern recognition, and decomposition. These skills can be taught through activities like coding, robotics, and data analysis, which can be integrated into various subjects, fostering a deeper understanding of computational concepts.
STEM Education and AI Ethics: a. A robust STEM education includes hands-on learning experiences and project-based assignments that engage students in scientific inquiry, engineering design, and mathematical problem-solving. This approach prepares students for future opportunities in AI-related fields and other emerging industries. b. Teaching AI ethics involves discussing the ethical, social, and legal implications of AI applications, such as privacy concerns, algorithmic bias, and the impact on the job market. By engaging students in debates, case studies, and group projects, educators can encourage critical thinking and responsible decision-making related to AI and technology.
Soft Skills and Emotional Intelligence: a. High school education should prioritize teaching communication, collaboration, and problem-solving skills through group projects, presentations, and class discussions. These experiences help students develop strong interpersonal skills and the ability to work effectively in diverse teams. b. Emotional intelligence can be cultivated by teaching students self-awareness, self-regulation, empathy, and social skills. This can be achieved through activities such as role-playing, mindfulness exercises, and reflective writing, which help students understand and manage their emotions and develop empathy towards others.
Adaptability and Lifelong Learning: a. Encourage a growth mindset by teaching students that intelligence and skills can be developed through effort, perseverance, and learning from mistakes. Celebrate progress and effort rather than just outcomes, fostering an environment that values continuous growth and improvement. b. Provide opportunities for students to explore new subjects, technologies, and learning methods, which can help them become more adaptable and open to change. This might include project-based learning, online courses, or extracurricular clubs that expose students to a wide range of disciplines and experiences.
Interdisciplinary Learning: a. Design interdisciplinary projects that require students to apply knowledge from multiple subjects to solve complex, real-world problems. This approach encourages creative problem-solving and enhances students' understanding of the interconnected nature of various disciplines. b. Promote collaboration between teachers of different subjects, encouraging them to design cross-curricular lessons or projects that highlight the connections between their respective fields. This approach can foster a more holistic learning experience for students.
Career Guidance and Exploration: a. Organize career fairs, workshops, and guest speaker events featuring professionals from various industries, including AI and technology-related fields. These events can help students gain insight into different career paths and the skills required to succeed in them. b. Offer mentorship programs or internships that connect students with professionals in their areas of interest. These experiences can provide valuable guidance, hands-on experience, and networking opportunities, helping students make informed decisions about their future careers.
By emphasizing these key areas, high school education can better prepare students for the challenges and opportunities presented by the age of AI. Providing a well-rounded education that incorporates digital literacy, STEM, soft skills, adaptability, interdisciplinary learning, and career exploration will equip students with the tools they need to succeed and contribute meaningfully to society and adapt to the evolving workforce landscape in a rapidly changing world.
When I asked again a day later, I got a different response:
In the age of artificial intelligence, the purpose and focus of high school education should be to adapt to the changing landscape by prioritizing the development of human capacities, protecting human rights, and preparing students for effective human-machine collaboration.
Here are two specific examples to illustrate the focus of high school education in the AI era:
Emphasizing project-based learning: To develop skills that complement AI, such as critical thinking, problem-solving, creativity, communication, and collaboration, high schools can integrate project-based learning into their curricula. This approach allows students to work on real-world problems and collaborate with their peers to find solutions, encouraging the development of skills that will be valuable in the AI-driven workforce. For example, students might work together to design and build a robot that can perform a specific task or develop a machine learning model to predict environmental changes.
Introducing interdisciplinary courses: High schools can offer interdisciplinary courses that combine technology, AI, and traditional subjects, allowing students to explore the impact of AI on various fields and develop a well-rounded understanding of the subject. For example, a course on "AI and Ethics" could explore the ethical considerations of implementing AI in various industries, while a course on "AI and Art" could investigate the ways in which AI can enhance or create new forms of artistic expression. By offering such interdisciplinary courses, high schools can prepare students for a future where AI technologies are deeply integrated into various aspects of life and work.
In summary, the purpose and focus of high school education in the age of AI should be to equip students with the skills, knowledge, and mindset necessary to thrive in a world increasingly influenced by AI and automation, while also addressing their holistic needs to ensure well-rounded development.
Either way, not bad for such a new technology. The prompt I gave it was also short, very broad, and rather straight forward. The more detailed the prompt and the more advanced the system becomes, the better the results will become.
Imagine what it’ll have to say next year!
The New Skills
According to technology specialist and former Warner Bros executive, Justin Kahn, a skill that will almost certainly become commonplace, and that will almost certainly be in demand across many different industries, will be the engineering or design of the prompts we use to interface with AI. This must be coupled with a knowledge of the available AI technologies and how to use them together in order to problem-solve, innovate, and maximize efficiency, creativity, and results. Offer the AI a simple or poorly-constructed prompt, and the level of the response will be inferior.
An important part of this process is "priming" the AI before asking the intended question. In a similar way that the AI is "trained" on a large data set, the user must "train" the AI before asking it a question. This involves first pasting (or typing) in information along with the instructions for the question, and there are many interesting and even creative ways to do this. (I will demonstrate this in a new blog post here, since this post is chiefly about education.)
Communication, language, and expression therefore take on a new level of importance, not to mention the fact that each subject area has its own ‘language’, and these will need to be leveraged in order to access quality and meaningful information. We must also be able to evaluate the quality of the AI's response, and repeat the process, if necessary, with a more effective prompt. In such a landscape, the skills of proof-reading and fact-checking take on a whole new level of vital importance as well.
Limitations Of The New & The Old
Of course, the ChatGPT tool has not yet reached its potential; these are very early days. ChatGPT is a large language model. This means that it synthesizes plausible sounding text responses based upon the input it has been fed, and upon which it has been trained. It will therefore have more difficulty with open-ended questions or those outside of the general knowledge base. It is not reasoning about things or even understanding the questions it is being asked. It is not even accessing the internet unless it specifically has a plug-in mining real-time internet data. But this level of connection is already arriving and will soon be standard fare.
In the meantime, though, the early versions of the technology often return results that read as "very convincing nonsense," as my friend and gifted UK software engineer, Innes Anderson-Morrison, calls it. And this is true. He contends that "ChatGPT is very convincing at sounding human, but as far as I can tell that's the goal: sounding human, not being a reliable source of information.” He continues that, "it's a very interesting time and we need to be keeping these developments in mind for the impact they are having, but the main issue is how people interact with tools like this and how we teach them to be critical of their output." Even though the quality of the responses will improve over time, two of the vitally important skills around using this technology must therefore include the ability to proof-read and fact-check information, as Justin Kahn suggested above.
If our educational system does not produce graduates with, at the very least, these types of new and necessary skills, it is no longer serving us and must be changed. Immediately. This will take courage, most of all, on the part of parents and educators. I think students have been waiting for a new approach, and they will embrace the shift far more readily than the adults around them.
It might even make education fun again.
Collaborative, Interdisciplinary, Project-Based
Before we conclude, let me add a few human-generated thoughts to ChatGPT's point #5 above regarding collaborative, interdisciplinary project-based learning.
Interdisciplinary study is not only an important way for students to expand their exposure, thinking, and problem solving abilities. More significantly, throughout the history of science, major breakthroughs have repeatedly been made from outside the discipline. This is because outside thinking has not been constrained by the accepted dogma within the field (which is an unfortunate consequence of human survival instinct). In fact, how could any accurate understanding of the universe be attained from within any one discipline. Nature contains so many disciplines, whose concepts flow one into the next, interwoven, with no clear separation. There are no contour lines in nature showing where physics ends and biology begins, nor where the geological rock cycle stops involving chemistry. These disciplines all overlap and have a bearing upon one another. Education and academia should reflect that ethos far more than they have done to date.
In the out-going paradigm, it has not been easy for teachers, for example, to collaborate with their colleagues in real interdisciplinary study. Firstly, they have existed under a time constraint to finish their curricula. Secondly, it involves additional preparation, not to mention coordination between teachers who are busy and, more than likely, over-worked. And some just want to do what they do — what they have always done — without interference.
But things are changing now in a way that will see those left behind who do not join the inevitable march of progress. This is especially true in education. School needs to produce humans who can explore, innovate, and achieve next-level things using next-level tools.
To use next-level tools, we must nurture a next-level vision.
And, ironically, this new mode of learning may well show itself to be not only more effective but also more enjoyable, for the teacher, as well as the student.
The Whole Discipline
As we move into this new paradigm, one mindset of the old system that we might be able to release from our desperate grasp is the notion that, in every subject they take, students must achieve mastery over the entire disciplinary area: We start at the beginning of the book at the beginning of the year, and we finish it in time to study for the final exam before the end of the year. And gaps in their knowledge must be avoided. In actuality, we may no longer need that focus because, moving forward, AI will be able to do most of the heavy lifting on that front for us.
Real education happens when a person engages with a direction of learning or endeavor and pursues it with interest and passion to whatever degree is necessary to achieve their intended goal. And to satisfy their curiosity. Such a pursuer of knowledge will almost always be motivated to achieve mastery of the area of their passion anyway, but it may no longer be appropriate to foist that expectation of mastery upon everyone passing through to glean their value for amongst the breath and depth of the subject. Real education is self-motivated and self-directed. It is time for us to free it from outdated constraints and to allow it to continue ‘marching off the map’ (as Tim Elmore's book of that title suggests).
That is not to say that breadth and depth are no longer important in education. Far from it. Breadth is necessary for perspective, and depth is necessary for specialization. But it will now become possible to make significant progress in an area, despite significant gaps in one’s own knowledge base, because we now have AI as a wing-man. (It will be no different to my using an image-generating AI, like Midjourney, to create gorgeous images when I have very limited artistic and graphic design abilities. And the same goes for you.)
Many times in my own research, I have come across concepts with which I am not familiar enough. I have then either had to go and consult with someone who knows the subject better than I, or I would have had to go and learn about those concepts myself in order to make progress on the problem I was working on. Sometimes the difficult concept is in the same discipline as the problem, but other times it is in another. (Although, no problem is really restricted to only one discipline.) If I were first required to master all of the disciplines involved before beginning my quest for answers, I would make far less, if any, meaningful progress.
What I was passionate about, in my research, was the endeavor, the question — the project. And furthermore, the world is one of challenges and problems that we must work to overcome through collaboration with others.
PICT Preparation For Life
It therefore seems that the logical way to approach education, in order to prepare the young to participate and contribute in this new world, is through Project-based, Interdisciplinary, Collaborative learning that leverages AI Technology to the full. This 'PICT' approach need not occupy 100% of student learning time, but it should certain occupy a significant portion of it.
A collaborative project can take one through all manner of disciplines, in ways that cannot necessarily be predicted at the start of the journey, and for which we could therefore never really fully prepare. But it requires all of the skills that represent an effective education: the ability to think critically and work towards effective solutions to real-world problems. Project based learning is therefore an excellent and effective way to get students pursuing a learning objective, in which pursuit they will be taken on an unexpected and fluid journey of discovery, one that should not be constrained by artificial or obsolete boundaries. And it will require the relaxation of many of the traditional boundaries between subjects and into designated time-slots.
In my opinion, the goal of project based learning in a modern, interdisciplinary environment is to develop the student’s ability to use technology to think critically and to problem solve. Technology can and should be fully leveraged to the degree that it is available, because upon this powerful foundation, we will be able to see (and reach) further.
(CLICK BELOW to see a few AI-generated ideas for PICT learning projects in several subject areas:)
The AI age is here.
There is no avoiding this world-changing trend. The power of AI tools will continue to grow, and they will help to accelerate the rate of human technological advancement.
All we have to do is know how to ask the questions! And that requires a new type of education.
The 2 images above were generated with an AI graphics tool called Midjourney, and using the prompt:
"Highly detailed high-resolution ultra-realistic photograph of high school students learning collaboratively in a futuristic setting containing futuristic glass monitors and holographic projections. --ar 16:9 --v 5"
Note: In the image on the left, the imperfections of the output of the AI can still be seen. Two examples include the girl in the background whose arm is cut off and the girl on the far left whose facial features do not look natural.
CLICK HERE to listen to a conversation with Dr. Stephen Wolfram on his plug-in for ChatGPT, and its exciting implications.
CLICK HERE to visit a growing "wakelet" repository of AI tools for students and teachers - brought to you by livingmaths.com.
For more from this author on the subject of emotional intelligence, suggested as part of module 3 above, see Evolutionology and the more detailed The Animal In The Mirror (coming soon) for more information.