Robotics, artificial intelligence and machine learning are no longer the stuff of science fiction. It’s simply a given that AIEd is becoming a progressive game changer on the national education front – but to what extent, to what effect and at what cost?

Where better to start answering these topical EdTech questions than delving into the explorations of the recently published Educational-AI-tion Rebooted? report from global innovation foundation Nesta.

Drawing on data mapping, surveys and research, this extensive paper summarises a selection of AIEd tools currently on the market alongside responses from the general public, before considering the future route and impact of AIEd in schools and colleges across the UK by 2035.

It’s go-to data resources include Crunchbase, Gateway to Research and YouGov; its surveys were compiled by 1,225 parents of school-age children; and data was collected from 69 AIEd companies, primarily located in and around London. Before moving onto the report, I’ll state its definition of artificial intelligence in education, as this can be somewhat subjective:

“Computers which perform cognitive tasks, usually associated with human minds, particularly learning and problem-solving”

Educational-AI-tion Rebooted? kicks off with three categories of AIEd: learner-facing, teacher-facing and system-facing. Let’s take a look at each ‘face’ in turn.

Learner-facing artificial intelligence education platforms – like CENTURY and Mathigon – offer students very personalised learning journeys. Learners are enabled to work at their own pace, often in their own time and according to their own strengths and weaknesses. My home is already familiar with CENTURY as my Year 7 daughter used it to good effect as flipped learning before attempting her school’s rather cumbersome Google Slide science lessons during lock down; and she has continued to use CENTURY to boost her English Language skills over the summer holidays, targeting her weaker areas (symbolism and onomatopoeia) in response to the platform’s automated AI assessment. Although neither of these learner-facing platforms have yet been rolled out by her secondary school, as a parent, I personally find the automated assessments in CENTURY insightful in targeting gaps in my daughter’s learning. Mathigon is news to me, so I aim to trial that learner-facing AIEd tool shortly.

The second AIEd platform detailed in the report is teacher-facing. It aims to “reduce their workload, gain insights about students and innovate in their classroom”. A platform that can swiftly and accurately administrate, deliver and assess each and every student’s progress – freeing up time teacher time for creative, well being and problem-solving tasks combined with collaborative group work and authentic live experiments – sounds like an absolute bonus. ClassCharts, which uses AI to find the best seating arrangement for all learners in a classroom, and a bespoke teacher-facing tool called Ada used at Bolton College, which automatically responds to student enquiries, are current examples.

This leaves the lesser known system-facing platform in third place. Why? This AIEd platform requires lots of shared data (e.g. DfE, school workforce census and parental views) across multiple educational institutions. It’s possibilities are far-reaching: it could be used to organise whole-school timetables, target specific schools for full Ofsted inspections, share best-practice teaching strategies and help address behavioural issues. However, collecting and pooling big data is the real challenge here. A topical example is Third Space Learning‘s collaboration with University College London’s Knowledge Lab, in which machine learning algorithms built into Third Space’s AIEd platform showcase patterns that result in positive teaching outcomes – and share those insights to benefit others.

So, that’s a brief summary of the current state of play. My next blog will round up the growing potential of AIEd in shaping our futures.