AI didn't create the crisis in education. It revealed it.

Most serious conversations about AI in education circles the same questions. What should we still teach when machines can do the thinking? What kind of learners are we forming? What gets lost when the path of least resistance is always one tap away?

These, to me, are not new questions. They are the questions education has been avoiding for more than a decade.

In 2017, I presented a paper at the Oxford Symposium in Comparative and International Education (OXSCIE). ChatGPT was not available to me. Claude did not exist. Generative AI was not in most people's vocabulary. The disruption to education that smart technologies might bring was still theoretical, something on the horizon.

This is part of what I shared at the roundtable session:
Innovation by nature is a state of constant change. With fast-advancing smart technologies rendering many jobs redundant, it is imperative that education programs in schools adapt to meet the new requirements of the working world. While the traditional method may have been effective in instilling discipline and maintaining an organised society in the past, in the 21st century, uniformity and conformity will only serve to hinder the creative and dynamic growth of learners, producing a generation of compliant individuals who will lack intellectual curiosity and the vital skills to design solutions for the endless and ever-changing flow of global challenges.

And later:
The uncertainties we face are continuously changing. They cause feelings of discomfort and vulnerability, yet are a part of life and cannot be eliminated. Rather than being consumed by the fear of the unknown, humanity in the 21st century needs to become ever more resilient and develop the art of being clear in the face of uncertainty.

Reading it back almost a decade later, I recognise how little the argument has had to change.

Nine years ago and in the room at Oxford with other roundtable speakers in my midst, we discussed and noted that uniformity and conformity were failing learners. The system that we were familiar with was producing compliant individuals when it should have been producing curious ones. The capacity to be clear in the face of uncertainty was already the more important capability, well before AI made the uncertainty more acute.

From this perspective, perhaps AI did not create the problems we now speak of. The reason the AI in education conversation feels urgent now is not because AI is a threat, but because AI revealed that our system had long been optimised for compliance, a system that rewards conformity.

The crisis I see is not that AI is too powerful, but that we have perhaps not been prepared, nor been preparing children, for a world in which thinking carefully matters. And now, a tool that punishes the absence of that preparation in real time, has arrived.

____

For interested readers, you can find the abstract I had submitted for OXSCIE 2017 here. Roundtable 9, Abstract 9C.
https://www.oxfordcie.org/oxscie2017-roundtables