Stay with me. I want to offer a few considerations about what I consider the inevitable transformation of education by artificial intelligence, but to do so, I’m going to first invite you into my childhood and early college years for a moment. It might not seem related to AI, but if you bear with me, I promise to offer you a few important and incredibly relevant considerations, as well as an important challenge and invitation.
Mr. Bently was an extraordinary teacher. Life wasn’t always easy in my elementary school years. Many others faced far greater challenges to be sure, but suffice it to say that when I went to school, it was not easy to set aside worries and concerns from outside of school enough to get the most out of what happened in most of my classes. Nonetheless, when I walked up to the room to enter Mr. Bentley’s class, he consistently greeted me and every other student at the door. As he wished us each a good morning, he also paid attention to the little things and deliberately said something that made each of us keenly aware that he cared about us and noticed us.
During class, he applied that same care and attention to each lesson. He seemed to notice small shifts in facial expressions that hinted at frustration, fear, or confusion. Not that he always came to the rescue, but he had a way of showing that he noticed while encouraging us to persist with a challenging problem. He listened to what we said, noticed what we didn’t say, keenly observed our nonverbal messages, and clearly worked to help cultivate a positive learning environment, most of the time without giving hardly any direct instructions.
I remember many caring teachers, but Mr. Bently stood apart from the rest as I think about teachers who listened, observed, and adjusted accordingly with such care and skill. How much did he care and pay attention? Twelve years after being his student, I was going into the summer after my freshman year of college. I had a summer job, but on a whim, while driving by an insurance company in my home town, I decided to stop in and ask if they had any summer openings. The next thing that I knew, I was in a beautiful office, speaking with the branch manager. Impressed with my initiative, he offered me a job on the spot, serving in their call center, working lists of prospective customers. Using a simple script, I spent evenings calling name after name, introducing myself by name and asking if they had interest in reviewing their insurance coverage. If so, my job was to schedule an appointment between that person and an available agent.
I didn’t enjoy the job. Making the calls and talking to people was enjoyable enough, but the list that I used included some problems. First, some of the people that I called were already customers, and they were often offended that I didn’t know as much. The worst calls were when I would ask for a given person, only to find out that this person passed away in the recent past. Imagine calling a person, asking for the spouse, only to discover that the spouse died in a car accident the day before, resulting in audible sobbing as you struggled for what to say. Out of a list of a couple thousand names, I remain amazed at how many deceased people were included on that list (Note that this was in the early 1990s, long before current methods for such work). One day, working through a new list, I reached the “B”s and found myself calling a number and asking for a “Mr. Bently.” A woman answered the phone and as I said, “Hi, my name is Bernard Bull…” the woman stopped me. “Bernard? Bernard Bull?” I confirmed. “Oh, my husband will be so delighted to speak with you.” This was the wife of my 4th grade teacher, Mr. Bently!
Did you catch that? Twelve years after being in this class, the wife of Mr. Bently recognized my name in an instant, and when I spoke to him, his memory of even the smallest details about me were fully intact. In an instant, it was like I was walking into Mr. Bently’s class all over again, experiencing that incredible felling of care, recognition, and belonging. I felt noticed and important to someone else, and it felt amazing. I don’t think that I went on with the rest of my script.
It would almost sound like blasphemy to think that artificial intelligence could ever replace a Mr. Bently. A good part of me continues to believe that no non-human system will ever serve as a substitute for the incredible and formative experience of being noticed and cared for by a teacher like Mr. Bently. Beyond that, he was a master of listening and observation, and he used that to help me and countless others learn. A great teacher like him is truly gifted at the art (and science?) of noticing nuances in learners and responding accordingly.
Perhaps that sort of a deeply human and meaningful interaction is only possible between two humans. Yet, we are on the verge of an age when artificial intelligence is inching, or sometimes leaping, toward noticing countless nuances. Consider what non-human systems can extract from a single still image of a person, breaking down facial expressions into the various combinations of muscles and movements in the face, even noticing the development of some muscles over others, potentially hinting at patterns of expression and emotion over time. These systems are emerging that promise to detect lies, fear and anxiety, interest, confusion, and more. Imagine a system that demonstrates the same capacity to notice nuances in our posture, tone of voice, choice of words (spoken and written), online habits and actions over time, in person action and habits over time, or reaction to various stimuli and contexts, and our response to any other sensory experience in the world. We are already partly there. Consider this enhanced by the ability to interpret what is happening in a person on the basis on heart rate, brain wave, and eye dilation, blood flow to various parts of the body, and other involuntary physical responses; comparing all of these “data points” to a massive database in order to diagnose and adapt.
Does this seem far-fetched? Scan the news and you will find articles about AI detecting skin cancer better than doctors, AI that can determine sexual orientation through still images, experiments with AI lie-detectors for border control, and AI behavioral systems being used in schools within China? We are talking about technology that is already getting heavy use in finance, healthcare, political strategy, security, social media, and yes, education. We might not have systems in education that are as advanced as I mentioned in the last paragraph, but we are well on our way.
Consider what happens as we reach a time when such technological observation is combined with the most current research on knowledge and skill acquisition. An artificially intelligent cyber-tutor will constantly read, analyze, and adapt learning experiences to maximize learner interest and progress. As these systems advance, they will far exceed the capacity of any human to facilitate learning for large numbers of learners, even across time and place. This is the future of adaptive learning, personalized learning, as well as individualized instruction. It is, I contend, inevitable and irreversible.
Will these systems greet students at the door? Will there even be a door or a classroom? That is yet to be seen. Will they fill the deeply human need to be noticed and cared for by another human? Even if they can, I personally hope that we count the cost before going that direction. My life today richer because Mr. Bently noticed and cared, and I’m not ready to sacrifice that at the altar of artificial intelligence. At the same time, there is incredible promise and possibility with such technology, and I’m not ready to sacrifice that on the other altar of nostalgia and sentimentality. Rather, I like to think that we can join in co-creating a future of education where the best of these two worlds come together, creating deeply human and caring communities that are transformed and enhanced by carefully considered artificial intelligence systems.
In addition, there are many learning needs throughout life that are already less high-touch and we are fine with that. We turn to online video tutorials to learn a new skill, read books and online guides, opt for largely impersonal training, use educational apps, and blend our learning throughout life with a mix of learning environments and formats. Some are human-driven. Others are not. As such, those in the latter category as well as those areas where the human-driven learning is falling short are both prime candidates for disruption, or at least significant experimentation as we explore the possibilities, affordances, and limitations of artificial intelligence in education.
There is much that we don’t know about the future of education. There are countless trends and innovations that will come and go. Artificial intelligence is not one of them. It is here to stay. It will continue to grow. It will find its way into an increasing number of contexts, eventually transforming many of them. The question is whether we are going to do the good and important work of helping to shape that transformation in positive ways, or whether we will simply let AI take the lead through lazy thinking, naivety, technological fatalism, or something else. Getting informed and involved in the conversation now is your chance to be a co-creator of that future. Now is the time for quick, low risk experimentation, careful consideration, wise thinking, and wide-spread discussion. I offer this article as one way to help spark that conversation.