Me, myself and AI: are robot teachers in our future?

Charley Rogers finds that artificial intelligence is well and truly here to stay

Posted by Julian Owen | December 11, 2017 | Technology

• #robot-teachers
• #ai
• #artificial-intelligence
• #turing-test
• #wayne-holmes
• #wise-summit

Hands up if you’ve heard the exclamation ‘teachers will be replaced by robots!’ within the last year. I imagine quite a few of you have. This worry, and others like it, is often raised as part of conversations around artificial intelligence (AI) and, more specifically, its place in the education system. 

If 2017 had buzzwords, then AI is surely one of them. A recent article on AI from the BBC was titled ‘Hype, hope, and fear’ – a pretty accurate summation of the kinds of attitudes surrounding the topic at present. But how much do these attitudes accurately reflect the way that AI is being implemented in education right now? And what about the future? The recent WISE summit in Doha, Qatar, presented a great panel on AI, and how global views on these topics are developing with each new advancement. Many of the issues covered referred to how different education systems are reacting to what is often considered an imposition on teaching, and what the practicalities of AI in the classroom really are. Expert insights from this panel, along with insider views from educator and AI expert Wayne Holmes, provide a much more balanced view of AI in education than clickbait headlines would have you believe. 

Learning personalisation

One of the most prominent benefits of AI in education is the ability to create personalised learning models, allowing each individual student to access material in a way that suits them, and at their own pace. In a traditional classroom, with one teacher in front of between 30 and 300 students, this level of individualised attention is impossible; not only does the teacher not have time to address each student, but relying on one explanation of the material means that a one-size-fits-all approach is inevitable. Jingfang Hao, a science-fiction writer and founder of WePlan, spoke at the WISE summit about how she envisages AI transforming education. Her insight was that personalised learning is the most important thing to come out of the technology, and that implementing AI algorithms into teaching will give teachers more time to help students develop soft skills; something that AI is thus far incapable of doing. “We are introducing these tools to help the teacher,” she said, “not to replace them.”

At this point, it’s important to distinguish between AI and robotics. Speaking recently to Wayne Holmes, Lecturer in the Institute of Educational Technology at the Open University, I discovered that these two things are oftentimes categorised together, most of the time subconsciously. “I do a lecture about AI in education, and the first question I ask is ‘when I say artificial intelligence what’s the first thing that comes into your head?’” says Wayne. “And then my next slide is a picture of a robot, and I ask who thought that, and almost half the group will always put their hands up.” The problem with this, explains Wayne, is that focusing on robots that are “easy to get photographs of” can detract from the very real applications of AI including algorithms, machine learning and processing, amongst other things. So when we think of an Asimov-style robot takeover, we are, in fact, miles away from the actual implementation of AI in education. 

The people vs. AI

Although providing teachers time to focus on students that may need some extra help, and providing algorithms and data sets to allow for personalised learning and tracking of progress, using AI features cannot, as yet, replace the human element. Jörg Dräger, a physicist and Executive Board Member of independent foundation Bertelsmann Stiftung was also a member of the AI discussion panel at the WISE summit, and made the point that the use of AI in education is a positive step, and can indeed provide a “wonderful orientation tool in helping to personalise education,” but that we need to consider whether we are putting tech or pedagogy first. A pedagogical implementation then, is the most important element of educational development, says Jörg, and although the use of tech is something of a no-brainer in making the process of education more effective and efficient, it is ultimately a support for the human interaction that already exists within the classroom, rather than a replacement of it. 

So, we’re not going to see a replacement of humans with robot teachers any time soon? Almost certainly not, says Wayne Holmes. “Intelligent teaching systems is one of the biggest areas in AI,” he says, but although these systems do replace a teaching function, there are elements of a human ability to judge emotion and explain difficult concepts that AIs simply do not possess just yet. However, there is a pressure from Big Tech to advance AI technology into this sphere, warns Wayne. “If big tech companies have their way, then there certainly will be a pressure [to replace human jobs with AIs] and I think it’s important that we make people aware of these possibilities, so people can challenge it,” he adds. 

For Yao Zhang, Founder and CEO of Roboterra, a Chinese company providing buildable robotics kits for children, a national AI strategy is required in order to make the most of the technology in education, while maintaining human-centric development. Speaking at the WISE summit, Yao also addressed the Big Tech problem, saying that it is unlikely for this to be an issue within the education sector, since it is so segmented. It is more likely that there will be a “large number of service providers with specific knowledge and skills,” she says. In this case then, and considering Wayne Holmes and Jörg Dräger’s views on the importance of the human in the classroom, AI is starting to look a lot less scary. 

It’s all in the data

So, if AI is not about robot teachers and a vision of sci-fi-esque human subservience, then what is it? For education especially, it’s all about data. Personalisation, time-saving, and interactive classroom models all rely on the effective use of data, and that’s where AI comes in. The artificial intelligence in play in education right now is essentially very specific data analysis software, differentiating data sets, and interpreting them to allow for reports on student ability, individual pupil strengths and weaknesses, and so much more. Imagine a classroom full of students that are working on computers. As they complete their assignments, the computer logs how much time is spent on each task, which elements of the task each pupil puts most effort into, and so on and so on. This data is then collected and analysed by AI software to which the teacher has access, allowing them to view detailed analysis on how each student is performing with each subject or task, and which students may need extra help.

If you think of learning materials as a specific collection of data, software like this can also help with initially providing information in class. A teacher can enter data, or content, into a computer – or use pre-loaded content from available apps – and allow the AI to interpret the content and relay it to students, freeing up more time for the teacher to curate this content, and answer more intricate questions from students that the AI cannot source from the original data provided. AI could even help with assessment, providing the ability to collect a portfolio of work throughout the school term or year, rather than relying on somewhat clunky and anachronistic examinations. As Jingfang Hao stated during the WISE summit, knowledge is more about the brain’s processes than its content. Thanks to the advent of the internet, teachers are no longer unique vessels of knowledge, but rather curators and interpreters of knowledge, teaching their students how to find the information they need, and how to differentiate this from ‘fake news’, or dubious sources. Jörg Dräger agreed with this sentiment, making the point that part of a modern teacher’s job is teaching their students how and when to use the tech tools available to them, to learn. 

System reset

It is the system, then, rather than the existence of the human, that has been and will continue to be altered by AI. From using the technology to build a portfolio of students’ skills and experiences, to relieving teachers of manually compiling reports and other time-consuming but relatively straightforward tasks, AI can be harnessed to revolutionise education around the world for the benefit of both students and teachers. But fear not, you’re not going to have to Turing Test your teaching staff any time soon; the human is here to stay, and – at least for now – robot teachers remain safely in the realm of science fiction.