News

AI could consign educational traumas to history

Receive free Social affairs updates

The writer, an FT contributing editor, is chief executive of the Royal Society of Arts

I am suffering from PESD (post-examination stress disorder). This is a new condition — indeed, I have just invented it — but it afflicts millions of parents whose children are put through the psychological mangle of school examinations.

The trauma is, of course, not confined to parents. Pity the students each year facing trial by examination, followed by weeks waiting on the verdict. Little wonder there is an epidemic of wellbeing and mental health problems among the young.

In the UK, a terrifying one in five young people aged 17 to 24 were reported as having a probable mental health disorder in 2022. Of course, not all are rooted in education and examinations. But the correlation is strikingly high between the incidence of mental health problems and low educational attainment in young people. It can be seen, too, in surveys of wellbeing — as many as a quarter of young people leave education loathing, rather than loving, learning. It would be hard to think of a worse endowment.

This means those departing education with fewest skills are also least likely to engage in life-long learning. And it means the gaps in attainment in early years are likely to widen over time, becoming chasms in adult opportunity and income. Education can entrench inequality, rather than redress it.

What could be done to close those chasms? A good start would be to rethink the metrics of success. Currently, examination results and school league tables are paramount. But if the key arbiter of later-life success is learning attitudes and experiences, rather than outcomes per se, it would be better to target those directly by measuring pupil wellbeing.

That’s not as idealistic as it sounds. In the UK, the #BeeWell initiative puts young peoples’ wellbeing at the heart of measuring achievement — and how well education is designed. After a pilot in Greater Manchester, this initiative is being rolled out to nearly 10 per cent of pupils across England this year.

Alongside this shift in assessing the educational ends, we are overdue a rethink of the means. Teaching in most schools is built on a 19th-century learning model — a “chalk and talk” learning and examination-based form of assessment. As a technology, it is mainframe and largely homogenous.

It has had historical success in raising educational standards and suits some learning styles (including, most likely, FT readers). But for many, including the neurodiverse, the traditional approach is ill-suited to their way of learning. It generates fear and loathing, and sows the seeds of a life-long learning deficit. They need a very different model — personalised, practical, team-based and hands on.

To date, limits on resources have meant a personalised learning was not practical for most pupils. While learning is often adjusted in its pace and content to different needs, truly personalised learning — tailored in its aims and outcomes, and pedagogy — has remained out of reach. Until now.

This is because digital technologies, and artificial intelligence in particular, are freeing us from our 19th-century learning constraints. The booming global “edtech” sector, now worth well in excess of $100bn each year, is testament to that.

Estonia, often a digital first-mover, is developing infrastructure using AI to give teachers and students greater autonomy, enhancing pupils’ experience and sense of agency. Has this damaged education outcomes? To the contrary, Estonia is the highest-ranked country in Europe on the Programme for International Student Assessment (Pisa) scale.

Personalised medicine, attuned to our individual biologies, is fast becoming a practical reality. Regrettably, personalised education, attuned to our individual brains and circumstances, is less advanced. Implementing such a programme would be the most significant shift in education practice in a century. But the potential dividend, for individuals and economies, could be huge.

It could help rescue life-long learning from the realms of rhetoric — essential, when it is predicted that nine out of 10 jobs will demand a significant skills shift by 2030. It could stimulate a love of learning for millions of otherwise lost learners, elevating their life satisfaction and income. Neurodiversity, for too long seen as a liability, could become an asset.

The aim of teaching should be enabling young people to “think for themselves and act for others”, in the words of educationalist Sir Tim Brighouse. Our current model often has the opposite effect. Technology now puts fixing this within our reach. For our children’s sake (and their PESD-suffering parents), let’s hope it is soon within our political grasp.