AI in Education: Revolutionary or reductive?

AI has been a hot topic across a range of industries lately. One where it’s been discussed plenty is education: how - and should - AI be used in teaching? It’s a complicated issue. AI can potentially revolutionise education, but unchecked has the potential to be harmful. Across many areas of education, AI can lead to efficiency improvements, but there are ethical and security concerns among others. We’ll consider both sides of the coin here.

How can AI help?

We’re still in the early stages of AI, and as it develops so too will the ways it can improve education. For now, we’ll be looking at the present advancements education can make use of.

Accessibility

This is one area AI has certainly helped. It can be used to help ensure students with disabilities have the same learning opportunities as others. Speech recognition and natural language processing (NLP) are great examples: for students with hearing or speech impediments, AI advancements in these areas are game changers. OpenAI has already been trained on 680,000 hours of multilingual data, with the aim of creating software with human-level robustness for speech recognition. NLP advancements can be used to create captions instantly for the hard of hearing.

An innovative, personalised learning experience

AI is already being used in new and creative ways to create interactive learning experiences. Virtual and augmented reality apps use AI to bring theory to life, offering visual learning on a new level. Our own product, Luna, uses AI to generate flashcards from any learning material. These sorts of advancements offer new, better teaching pathways which accommodate more learners. AI can also adapt and learn each student’s specific needs. Tutoring systems that use AI can provide responsive content and customised feedback to each learner.

Takes care of the admin

From a teacher's point of view, AI can automate the pesky administrative tasks that take their time: grading, scheduling, and data analysis to name a few. This frees up their time to focus on the human side of teaching. For example, providing quality instruction and having personal interaction with each student to assess their needs - things AI can’t replace.

Another thing that AI can do is analyse huge volumes of data in moments. This can be used to provide trends and patterns on student performance, highlighting areas that might need attention or where the curriculum is failing. This will increase the ability of educators to make evidence-based decisions on teaching strategies.

What’s the downside?

For all the brilliance AI can provide in education, it’s a very new technology with emerging concerns. Over time these might be mitigated, but we need to be aware of them.

Is it really inclusive?

There’s no doubt AI has massive potential to create new ways to support disabled students. There’s also a risk of worsening educational inequalities that already exist, though. In its early days, AI is cutting-edge technology which comes with financial and logistical requirements. Many students may be unable to afford the AI software which could help them or lack the stable internet connection required. This could cause a deepening disparity and widened divide, leaving some students behind in the AI revolution.

Even those who can use AI might not benefit from it. It could be overwhelming for some: where do you begin with a chatbot that can be asked anything? How do you prompt it in ways to get the answer you’re looking for? Students with anxiety may well find this harder than more linear teaching methods.

Implementing AI in some institutions also requires collecting huge amounts of student data. Safeguarding this data is absolutely vital, and is something many institutions won’t have considered on such a massive scale. Modern security measures will need to be implemented in response to this.

Does it actually improve teaching?

Earlier I suggested AI will allow more human interaction between students and teachers. There’s a flip side to this, though: if we go too far with technology, will we remove this human side entirely? There’s an undoubted emotional side to educating - mentorship, empathy, and guidance are critical parts of a student’s development. Overreliance on technology might mean we lose this in a way that’s hard to replace.

AI is also only as intelligent as the data it’s been trained on. If it lacks data or has been fed biased information, this will be what students are educated with. We run the risk of losing the diversity that comes from the human experience of teachers.

The ethics of AI

Certain AI makes use of data in a morally grey way. Some software will track the behaviour and emotions of students to make decisions on their education. But is this not a breach of privacy and a form of over-monitoring students? Currently, there’s a severe lack of ethical guidelines around AI in all industries. Technology has advanced faster than policymakers can keep up with, and there’s a lag in protective measures for AI in teaching. Indeed, school leaders have warned AI is a ‘real and present’ danger to education.

Where do we go from here?

Whilst there are concerns around AI, they can all be addressed with careful consideration and don’t outweigh the benefits. There have been suggestions AI development be suspended altogether whilst rules are put into place, but this seems extreme. Funding through initiatives like the Disabled Students’ Allowance will help avoid a digital gap, and if used correctly modern data protection measures will mitigate security concerns. We’re also a long way from a point where AI could come close to replacing educators, allowing plenty of time to ensure the human side of teaching is protected.

Regardless of education, business will continue to use AI as a means of improving efficiency. Students may well need to be prepared to use AI in their work. Much has been made of AI replacing jobs, but the reality is those who can use AI might replace those who cannot. If we don’t integrate AI with education, students run the risk of being the ‘cannot’.

The fact AI can be used to provide tailored and efficient learning experiences for each student gives it inherent value. Age-old problems in teaching are getting new solutions by the day and embracing that is the way forward. Any new technology comes with risks: overcoming them is the route to progress.