Artificial Intelligence and the Future of K-12 Schools
Concerns About the Uses of AI in Education (Part 2)
“Nothing could be more absurd than an experiment in which computers are placed in a classroom where nothing else is changed.” (Seymour Papert; South African/American mathematician, computer scientist, and educator; 1928-2016.)
“As more and more artificial intelligence is entering into the world, more and more emotional intelligence must enter into leadership.” (Amit Ray; Indian author and spiritual master known for his teachings on meditation, yoga, peace, and compassion; 1960-.)
“… we have a lot of children in each class, sometimes up to thirty. There’s no way a teacher can manage them all. It’s too overwhelming, as there are too many children, and too many places a teacher needs to be all at once. A human just can’t do that. An AI [[AI-using computer assisted instruction system] on the other hand could. It could be everywhere in the classroom at once, managing every child at once. This would result in every kid getting as much attention as they needed.” (Parker Heintz, 5th grade student in a Portland, Oregon public school.)
Introduction
Notice that the third quotation comes from a fifth grade student. The world, she is a-changing.
This is a continuation of the discussion in the previous IAE Newsletter regarding concerns about the uses of Artificial Intelligence (AI) in our K-12 schools. In this discussion, as well as in the four earlier newsletters in this series, we note that AI-using computers already are firmly entrenched in our schools and in our everyday lives. These newsletters emphasize that such technology is a major change agent in our world, and that its high pace of research and development will continue for many years to come. AI already surpasses human capabilities in a number of areas, and these capabilities also will continue to increase rapidly. However, you may be relieved by my forecast that the development of walking talking robots that can mingle with and pass themselves off as human beings will not occur in the next decade. (Pretty sneaky, right? My forecast hints that I believe it will eventually occur, but likely not during my lifetime.)
I tend to become philosophical as I write my newsletters. Right now, although I am focusing specifically on AI-using computers in schools, I cannot help but think about the fact that education is a continuing lifelong endeavor, with much of a person’s learning occurring outside of formal schooling. We all know that schools provide an important part of a person’s lifelong education, but this time is limited in hours per day, days per week, weeks in a school year, and years of formal schooling. Thus, we must pay careful attention to how this precious time for formal schooling is used, and we must align this special time with each person’s hugely more hours of learning that is occurring outside of the school hours. I will explore this topic more in the next newsletter.
An Expanded List of Concerns
My list of AI-related concerns presented in this newsletter includes an enhanced version of items from the previous newsletter, as well as additional items. For each concern in my current list, educators who are developing and implementing the curriculum need to consider whether the current use of such AI-using computer technology in our schools is providing students with an education that better prepares them for responsible adulthood in our current and changing world.
This list is by no means exhaustive. My intention is to provide a representative sample of ideas that have occurred to me. I am sure that you can add to the list and quite possibly may not agree with some of my concerns. I have not attempted to order my list in what I might consider to be increasing importance. As they say, “beauty is in the eye of the beholder”.
All items on my list, and others that will occur to you, need to be analyzed in terms of whether they would represent progress in education. But, progress is a hard word to define, So, we will never have complete agreement on lists of suggestions on how to improve our schools. Possible meanings of the term progress are discussed in the next newsletter.
1. An Overall Problem
People in the U.S. today have a great deal of mobility in terms of where they live. Thus, a significant percentage of students experience living in two or often more school districts during their K-12 school years. This always has posed somewhat of a problem, despite the fact that there a considerable amount of uniformity of content in the primary school curriculum and in much of the secondary school curriculum throughout the country.
We damage that uniformity when we provide varying levels of computer capabilities and varying types of usage of these computer facilities from school district to school district. This problem is exacerbated as schools provide varying levels of computer capabilities for students to use in their homes, and varying levels of amounts and types of uses of the at-home computers for school purposes. A related problem that will persist for many years is the varying levels of computer knowledge and skills of teachers at all grade levels, and those of the students’ parents/guardians.
Perhaps even more important to the future of schooling is the continuing rapid pace of change in the many AI capabilities that are relevant to K-12 education today. This represents an ongoing challenge to curriculum developers, schools, and parents of school-age children.
2. The Use of Electronic Calculators in the Math Curriculum
What is gained and what is lost in math education as we decrease the emphasis on paper-and-pencil by hand calculations, and increase the emphasis on using calculators and computers? It currently takes considerable school time for a student to develop paper-and-pencil skills in doing arithmetic calculations. Might the use of calculators and computers allow us to spend less school math education time on this topic, and more on other aspects of math such as mental calculations and estimations, number sense, math sense, math modeling, and problem solving? This is an important research question.
It also is a course content question to be asked across all curriculum areas as computers with powerful problem-solving capabilities become more and more available not only to students, but to adults who need to solve such problems. From quite early on, this question has been a challenge to our math educational system, as we have long had computer systems that could solve essentially all of the types of math problems that students encounter up through the first year or two of college math. Which of these problems must they still learn to solve by hand, and which should they now be learning to solve with the use of computers instead?
My doctorate was in mathematics, and I have long been interested in these types of questions. Very early on, I wondered whether the use of electronic calculators in grade school might eventually lead to a substantial decrease in the number of students who take the more advanced math courses in high school and college, or might the opposite outcome occur. Might students find that the math they are encountering in more advanced courses is so much more interesting and fun, they happily will pursue advanced coursework in math and in college degree areas that make extensive use of such math?
3. An Extension of the Item 2 Idea to Other Curriculum Areas
I see two aspects of this question, the first one from a math point of view. Math is a useful tool in most academic areas of study. Many teachers are loath to make such use of math in their courses because of the large variations in students’ calculation and other math skills, and because the teachers themselves do not want to spend their valuable class time helping their students learn to do math.
The second aspect consists of considering other computer tools that are useful in many different curriculum areas. For example, the use of computer graphics is now a routine part of the graphic arts, and graphic arts now are used routinely in the creation of multimedia documents—so learning to make use of such tools now has become an accepted part of learning to develop and write multimedia documents. To what extent do we want students to be dependent on computers for completing graphic arts activities that formerly were done only by hand?
More broadly, we certainly want students to learn to read in an interactive multimedia environment. Do we want them to learn to write in this environment? Students certainly can learn to do this, even at the elementary school level. But, think about the challenge to teachers in all curriculum areas that currently expect their students to produce written documents.
The point is that, because computer applications in computer graphics, spreadsheets, and databases are now so important and widely used, perhaps all students should begin learning to use these tools while still in elementary school. Suppose only some elementary schools decide to do this. Then teachers in all of the schools that receive graduates from these elementary schools will be faced by students who have these skills and who also have an expectation of routinely making use of them in the upper grades.
4. The Decrease in Emphasis on Cursive Reading and Writing
I see two aspects of this decrease in emphasis on cursive reading and writing that some schools are implementing. First, students are growing up in a world that predominately still makes use of cursive writing. So, an elementary school’s decision not to teach cursive reading and writing ill prepares students to deal with this mode of written communication that may well be expected by their teachers at higher grade levels, and a skill that is still a routine part of the adult world.
The second aspect consists of the benefits students gain by learning to make effective use of the grammar, spelling, and other writing aids that a word processor provides. A downside it that students may come to depend on such writing aids, and will be disadvantaged when they are not allowed to use them when taking tests.
As an old timer, I am somewhat disturbed by seeing some of my grandchildren struggle with reading cursive. That example illustrates some of the challenges that schools face as they consider increasing the use of computers in writing. However, there is much more to this issue. A good word processor has AI-based spelling and grammar checkers. This is a type of built-in Computer-assisted Learning. Such a system also provides good access to a list of synonyms and to a thesaurus.
5. Our Current Slow Progress in Developing Authentic Assessment of Computer-using Students
This is both a local and a national problem. In the adult world, employees who have need of computer access to do their job will have such access and are expected to use it. In addition, many adults with computer knowledge and skills make routine use of these skills in their day-to-day lives. Thus, as students gain increased computer knowledge and skills, we need to assess their progress to determine how well they are doing—because such knowledge and skills will be very important to these students in their adult lives.
We already have come to expect students to make use of the Web as they complete assignments, but currently most teachers do not allow such access when students are taking tests. Traditional schools have a severe lack of open computer, open connectivity in assessments. National, state, and regional tests lack this authenticity. These are all problems that face our educational systems.
6. The Replacement of “Conventional” Library Materials by Online Library Materials
There is considerable difference between the knowledge and skills used as one searches in a card catalogue and browses the library stacks, versus searching on the Web. Conversations with a skilled school librarian, and formulating search strategies based on such conversations, is a valuable part of education that is disappearing rapidly from our schools. Classroom teachers in each subject area must now teach the library research skills related to accessing and using the types of content relevant to the subjects they are teaching. Not only that, such teachers have an added concern of more widespread plagiarism resulting from sudents having easy access to the Web. (Of course, many teachers now make use of the AI-based software that searches a student’s paper for signs of plagiarism.)
Certainly there is a difference between learning to browse the library stacks and learning to carry out a Web search. Personally, I like to make an analogy between browsing library stacks and browsing in a retail store, where I often stumble upon a product that grabs my attention. When I do a Web search, in essence I am using a computer to search though billions of documents and bring to my attention a few that its AI determines might interest me. While this is an increase in efficiency and also searches a great many more documents than would be available in a local library, I miss the opportunity to browse the library shelves for unexpected discoveries.
The placement of items in a store is designed to increase sales. A search engine such as Google makes its decisions of what to display to me based on its knowledge about my previous searches and also about usage patterns from other people who have done somewhat similar searches. Many people are concerned by the collection and use of such information about their searches by the computer in its decision making, and of the subsequent sale of such information.
7. An Increased Risk of Students Being Presented with and Using Fake News
Fake news is defined quite broadly as content that is heavily biased or deliberately falsified. Here we are mainly concerned about fake news available through the Web and other resources readily available for students to read, listen to, or view. In the past, such materials in schools and classrooms were carefully screened by librarians and teachers. Even then, of course, children received (and nowadays continue to receive) considerable information from family and friends that may be strongly biased and/or just plain incorrect.
This presents an interesting challenge to teachers and to all other people who interact with children. It is part of being a child and growing up to have to deal with inconsistent stories (fake news) coming to them throughout the day. I believe this is a topic that needs to be addressed openly in schools. Certainly, as the reading skills of students reach a level at which they can access and read Web documents, they should be learning about fact checker websites. They should learn that it is perfectly appropriate for people to have differing beliefs, but it is important to pay attention to factual information that is based on careful data gathering, analysis, and peer review.
8. AI Is Now Routinely Used in (Very Addictive) Computer Games
Some of the earliest computer games were designed to be both entertaining and educational, and the number of these games continues to grow. Many of these are very useful educational activities often presented as simulations or in other game formats. For example, consider an airplane flight simulator designed to help a person learn to fly a plane. It is fun to learn some of the aspects of flying a real airplane, even if one decides to not become a pilot. And, of course, flight simulators are an important aid to training pilots of airplanes and rocket ships. Or, consider a computer simulation of a scanning electron microscope. Since the human interface to such a machine is a computer program, a user cannot readily differentiate between using a simulator and using the actual microscope. Simulations of this type are used routinely in training radar operators, air traffic controllers, and so on.
However, many of the so-called educational games are very short on educational value and very long on entertainment value. Years of research on designing these edutainment games has produced games that often can be increasingly addictive. This type of addiction is different from addictive drugs, but tends to have some of the same ill effects.
I personally have experienced this form of addiction in my own life, and I have learned to deal with it in a variety of ways. One of my observations is that my brain’s pleasure center (or, the part of my brain that gains satisfaction with what I am doing and accomplishing) does not differentiate between the pleasure that it gets from my playing a game and pleasure from my writing insightful paragraphs such as the ones I am currently writing.
Students now are routinely exposed to and play such games, many of which are in some sense brainwashing devices. What do we want students to learn about this type of addiction? Researchers and developers will continue to improve the capabilities (the entertainment and addictiveness) of such games. I strongly believe that schools should provide students with explicit instruction about this topic, and what to do if it becomes a personal problem for them. Free professional counseling should be made available to students who need help in dealing with such an addiction.
9. Learning to Learn in Highly Interactive Intelligent Computer-assisted Learning environments (HIICAL).
I have used the term HIICAL in a number of previous writings, but it is not yet widely used by other people. For me, the term emphasizes the role of AI in modern Computer-assisted Learning (CAL).
My concern is that we will underestimate the magnitude of this learning challenge faced by students and schools as the availability and use of such instructional materials continues to grow. We certainly have seen this as millions of children began to be home-schooled because of the Covid-19 pandemic. For the most part, students, teachers, and parents/guardians were ill prepared for such a sudden and massive change in teaching and learning.
HIICAL researchers and developers have had more than fifty years to hone their skills and learn to market their products. We have long known the value of providing students with individual, well-qualified tutors (Bloom, 2021, link). Now we have HIICAL that can do much of this same tutorial work with students, and so face the challenge of integrating the effective use of both human tutors and HIICAL.
The future will bring us more and better multimedia HIICAL materials designed specifically for use by children at the various grade levels and in the various subject areas. (And, of course, when it is useful to the learning task at hand, the materials will include virtual reality.) We are used to the idea of having different reading groups in a class. With sufficient high-quality HIICAL, it will become possible to provide HIICAL across the curriculum and across the grades. In the years to come, our schools will need to decide the extent to which it wants to incorporate this level of individualization offered by HIICAL.
While the financial challenge of providing the needed equipment, connectivity, and multimedia HIICAL materials is large, it is clear that these barriers can and will be overcome. We face other, more daunting challenges. For example, we certainly want students to have access to and also learn to make effective use of these facilities and aids to learning both at home and at school. But, such a learning environment is alien to most parents. (Contrast this with reading and writing literacy in at least one natural language, which is common in U.S. homes.) Teacher education will be a related and ongoing challenge as significant improvements in the capability of HIICAL and other uses of AI continue to occur from year to year.
10. Brain Implants and Genetic Engineering of Humans
“We’ll never have true AI without first understanding the brain.” (Will Douglass Heaven, currently senior editor for AI at MIT Technology Review.)
Humans have had thousands of years of experience in the genetic engineering of agricultural plants and animals, a growing field for scientific research. AI now plays a major role in both the research and in the implementation of this research. It is quite likely that our schools may soon be faced with learning to deal with an increasing number of genetically engineered children.
A somewhat related, but likely much larger problem, is that of brain implants. We are used to the idea of implants as replacements for various parts of a person’s body. But, the direct implant of an electronic human-to-computer interface is new. Research on proof-of-concept has been occurring during the past few years, and a highly computerized robot has been built that can actually carry out such implants (Kay, 2/2/2021, link). Of course, direct interfaces between computers and a person’s brain have been part of science fiction for many years. In such stories, computers can download a natural language into a person’s brain in a modest number of hours. That idea is still wild-eyed speculation!
Final Remarks
Since the initial development of reading and writing, and of arithmetic that makes use of reading and writing, education as a whole and our schools have never faced a challenge as daunting as that posed by current progress in ICT, AI, and technology based on their rapidly increasing capabilities. Continuing business as usual, with its glacial rate of change in our schools, does not suffice.
We know that business, industry, medicine, politics, and other major aspects of our world are making large and concerted efforts to adjust to and take full advantage of these advances in technology. It is imperative that our educational systems do the same.
References and Resources
Bloom, B. (2021). Bloom’s 2 sigma problem. Wikipedia. Retrieved 3/25/2021 from https://en.wikipedia.org/wiki/Bloom%27s_2_sigma_problem.
Kay, G. (2/2/2021). Elon Musk says Neuralink could start planting computer chips in human brains within the year. Insider. Retrieved 3/25/2021 from https://www.businessinsider.com/elon-musk-predicts-neuralink-chip-human-brain-trials-possible-2021-2021-2.
Kurzweil, R. (11/3/2019). The future of intelligence, artificial and natural. Innovation Global. (video, 63 minutes.) Retrieved 3/23/2021 from https://www.youtube.com/watch?v=Kd17c5m4kdM&feature=youtu.be.
Nordrum, A. (2/24/2021. What does progress mean to you? Technology Review. Retrieved 3/23/2021 from https://www.technologyreview.com/2021/02/24/1018424/technology-progress-growth-future-innovation/.
Roschelle, J., Lester, J., & Fusco, J. (eds.) (2020). AI and the future of learning: Expert panel report. Digital Promise. Retrieved 3/25/2021 from https://circls.org/reports/ai-report.
Author
David Moursund is an Emeritus Professor of Education at the University of Oregon, and editor of the IAE Newsletter. His professional career includes founding the International Society for Technology in Education (ISTE) in 1979, serving as ISTE’s executive officer for 19 years, and establishing ISTE’s flagship publication, Learning and Leading with Technology (now published by ISTE as Empowered Learner). He was the major professor or co-major professor for 82 doctoral students. He has presented hundreds of professional talks and workshops. He has authored or coauthored more than 60 academic books and hundreds of articles.