AI in education

This post is the first in a series which focuses on the role of generative AI in UK education. Starting September 2023, it aims to chart my journey through this rapidly changing area as I consider the practical implications of AI for both teachers and children. Please note, older posts may no longer be relevant. Delve in and please do get in touch with any thoughts.

The World Economic Forum recently published that 90% of teachers have never had any training or advice on the impact of AI in the classroom — which is hardly surprising, considering ChatGPT launched under a year ago and we teachers have a pretty hectic time of it. So some of my work this year is an attempt to bridge the gap between EdTech and busy teachers. My plan isn’t just to talk about it hypothetically, but to trial and share findings: to work out where AI tools might actually help and where the trip hazards lie.

This work will (at least initially) rest on five main premises:

To break these down a little:

AI probably won’t signal the end of teaching as we know it

In the clearest guidance report I’ve seen on AI in education so far, UNESCO clearly states that generative AI won’t solve educational problems, and in fact is likely to add to them. The pandemic has demonstrated how vital it is that teachers and pupils share real space in real time and that motivation, confidence and progress stagnates when these aren’t in place. For many of us, the prediction of AI bots replacing human teachers and online platforms replacing classrooms just isn’t realistic.

…but the right tools could streamline educational processes

The low-hanging fruit are likely to be lesson resourcing, adapting learning materials for different learning and language needs, marking, data analysis and communications. Many teachers are already using large-language models in these areas, as Rebecca Allen discovered using TeacherTapp responses. Over the next year I will be collecting and trialling as many tools and approaches as possible — watch this space.

A personalised learning tool?

The dream of all students having access to a free, omniscient and friendly tutor, who can learn their areas of learning need, is undoubtedly closer. With the right prompts, you can get exactly what you need, explained in the way that works for you. We might see generative AI work for writing the same way that calculators have worked for maths: while it won’t supersede the need to learn basic writing components (vocabulary and sentence construction), it will allow students to (when appropriate), leapfrog these clunky construction phases into higher order thinking of argument, expression and debate more quickly. AI at everybody’s fingertips can avoid wasted time of irrelevant instruction and optimise the progress learners can make.

…but this will widen the achievement gap

All the above depends on two crucial and inextricably linked things: pupil motivation and reading ability. Large-language models are — by definition — text based. If you struggle with reading, you are locked out. If you aren’t intrinsically motivated, AI tools will undermine your ability to learn. This, of course, makes AI more damaging the younger you are, when reading confidence is hugely influenced by socioeconomic background. The very real danger is that open-source educational tools which rely on LLMs seem democratic, but in practice are anything but.

Reading and digital literacy

The only way to combat these dangers are to throw everything at reading and to augment our view of digital literacy. Pupils need to be taught to decode and comprehend fluently and confidently, probably before they are given unfettered access to AI tools. Age and stage for AI introduction will need to be carefully considered — and this is made even more complicated when there emerges a technological chasm between home life and school life. Alongside a solid focus on reading instruction, we need to ensure children are aware of the sources of information online, and how this is greatly complicated by the language models AI tools rely on.

Elisabeth Bowling is Vice Principal at the Inspiration Trust. She tweets at @elucymay.