University College London professor Wayne Holmes often shows a photograph of a baby alongside a chip in his lectures. He then asks the audience: If we are going to continue to base education solely on transmitting knowledge, why don’t we just put all of Wikipedia’s knowledge on a chip and implant it into babies? It is something that can be done, he says, in the relatively near future. So, he insists, if education is only about transmitting knowledge, let’s put the chip in our brains and close the school, because it will no longer be necessary. “Obviously not. You go to school to learn knowledge, but also to be a social being, to understand how to interact with others, how to develop critical skills, creativity… ”, explains Holmes (Ontario, Canada, 62 years old). An expert in artificial intelligence applied to education, an advisor to international organizations such as UNESCO, he participated in the latest edition of the Enlighted forum, which brought together 150 world experts in education and technology last month. Shortly before, he attended this newspaper by videoconference.
Question. Do you usually insist in your presentations that teachers should be involved in the debate on artificial intelligence in education? What will happen if they don’t?
Answer. In the field of education technology, we have specialists, technicians, engineers in artificial intelligence who are very good at what they do. They are very well trained. They are experts. And, when it comes to seeing where they apply their experience, they think: ‘Education? I understand this from education. I went to school, my children go to school, I understand education ”. But actually, they have very superficial knowledge, so what they’re trying to do is automate things, but things they shouldn’t. For example, in the last almost 50 years, there has been a community of researchers in artificial intelligence applied to education who have achieved great things. But in the last eight years, the academic field has declined and now companies are doing it. And what is happening is that the most common application right now is what is called “smart tutoring system”. With this technology, the child sits in front of the screen and interacts with the computer, because the computer knows more, of course. And the computer takes the student along a path that adapts to his achievements and difficulties to lead him to the correct answer. The problem is that, somehow, they claim that this is better than a teacher. But is not; that’s absurd. They can be useful, according to, to complement the teacher’s work with homework or with tasks to do on the bus on the way to school, for example.
P. And what should you be working on then?
R. I think most teachers are very good at what they do and that education and learning is a social activity where one plus one makes three. We learn together. And the problem is that these technologies separate us. Have you seen those absurd photos of 30 students staring at the individual computer, ignoring the child to their right and left, ignoring the teacher? And is that supposedly better? So what could we do? We could focus on supporting the teacher to be stronger. Who is researching a virtual artificial intelligence exoskeleton for teachers? Have you seen these exoskeletons that help lift heavy weights? That’s what I mean, why don’t we have one of those virtual ones for teachers? I will give a concrete example in the field of collaborative learning. We know that when students learn together they do better than alone. But we also know that it is very difficult for a teacher to configure that learning. How to make the appropriate groupings? Thus, a very simple tool would be one that makes those groupings [en función de sus características, sus resultados]. Although the teacher can move or change it later, this could save you 10 minutes in each class and facilitate that collaborative learning.
P. Is there anyone working on that kind of technology?
R. Not that I know.
P. These are the applications that, in his opinion, artificial intelligence should have to improve the way of teaching. But should it also have an impact on the content of teaching?
R. First of all, I think we should help all of our students understand what artificial intelligence is. In my opinion, what makes it qualitatively different from the first technologies is that it seems to do things by itself, although in reality it does not. It is always programmed by a human and we need all of our students to understand what that means, what impact it can have [las decisiones de esos seres humanos] in our jobs, in our lives. And it’s not just about the technology. How can artificial intelligence be made useful? What are the human consequences? And we should teach it in different subjects. For example, in Music, we could show a piece of music written by a human and another written by a bot and ask the student: what do you think? What are the challenges that this poses for us? In Language, the same, but with a poem written by a machine … In general, we have to go beyond the model in which the teacher gives knowledge to the students. Because computers are very good at maintaining knowledge correctly. And accessing that knowledge through a Google search is very easy. So, owning knowledge is no longer what we should be focusing on, but helping our young people understand how to evaluate knowledge.
P. These ideas about the radical changes that technologies bring to schools have been around for many years, decades. First, with audiovisual media, especially with video, then with computers and the Internet. But the truth is that the school hasn’t changed much. Do you think artificial intelligence is going to transform it this time for real?
R. There is a book, Teaching Machines: The Story of Personalized Learningby Audrey Watters (Teaching Machines: The History of Personalized Learning, MIT Press, 2021) that reviews educational technologies for 85 years. And it turns out that many of the statements we hear today about artificial intelligence, about saving work for teachers, about personalized teaching, were already proclaimed 85 years ago. And here we are. However, do I think technology can be useful in school? Yes, of course. But not to do these ridiculous things that people keep demanding. Will artificial intelligence change schools? No, at least for many years. During the pandemic, everyone said that the situation would change education forever, that it would change exams forever. But here we are, 18 months later, and we’ve gone back to school to do the exact same thing.
P. So, do you think that when the pandemic finally passes, we will return to where we were before?
R. I think so. There were so many optimistic people [con el cambio] and so many organizations, like the OECD or UNESCO, organizing big events about the opportunity that the crisis presented to improve education … I praise that enthusiasm and the fact that people are trying, but nothing really happened. Absolutely nothing.
P. Now a lot of money is coming to Spain from the European Union to digitize the education system [el proyecto de Presupuestos Generales del Estado para 2022 computa algo más de 1.400 millones de euros]. What would you say to the administrators who have to design the programs to spend all that money?
R. The first thing is that I believe that digitization is important, but it is not a solution in itself. It is true that there are some things that have to be guaranteed, such as the infrastructure, which must be adequate, for example, if we want children to be able to work at home and connect. The other issue is training, specific knowledge. In the United Kingdom, during the pandemic, the Government gave laptops to students that they did not have, but they could have thrown them in the trash, they were useless, because they did not give these children or their parents training or support. They also did not ensure that the laptops had the proper materials installed, nor did they offer computer support. It is an entire ecosystem that has to be prepared. And the training of teachers is here again the key. So the key to spending the money is that the part dedicated to buying technology should not be more than half of the entire budget. Because the other half must help people to specialize, to communicate with each other. And if you put the technology on that stage, maybe you have a chance to make it work. But if you just put it in school, you’re going to get some really skilled teachers to do something amazing, but the remaining 98% won’t. That is the tragedy. And we’ve seen it before.
P. Do you have any advice for that teacher who is getting the technology but no one has really explained what to do with it?
R. First of all, my heart goes out to him because it is a very complex situation. And part of the problem is that, to begin with, there is no evidence, good evidence, that many of the technologies actually do what they promise. But even for the small number of cases where there is evidence, how does that translate to a teacher? It is necessary to translate technological language into educational language, it is necessary for teachers to work together to build communities and to share the experience of using technology. So, to that particular teacher, I would say: hold on to the ones you already know and build it up gradually. Be confident in what you are doing and then try other things. Talk to other teachers. Search online forums, but don’t believe it marketing, because its not true.
#Artificial #intelligence #expert #Wayne #Holmes #Research #exoskeleton #teachers #smart #tutorials