AI and EdTech
EdTech and AI for learning: time to reassess?
Digital technology has become an everyday part of school life. In the DfE’s 2025 survey, more than nine in ten state schools had laptops available for pupils (90% of primaries, 94% of secondaries) and 87% of primaries had tablets.
Much of this change happened rapidly, particularly during the pandemic, when schools were forced to adapt at speed to keep children learning remotely. Since then, evidence about the impact of educational technology – EdTech – on the learning process has evolved, raising some important questions.
A growing body of research shows that more technology does not mean better learning or greater digital skills. On the contrary, increasing time on screens often correlates with weaker foundational knowledge and core skills.
We believe it is time to examine this evidence and to respond with appropriate changes. Legislation that exists in other areas, such as food, medicine, and toys, to ensure products are not harmful for children, does not yet exist for EdTech. It is therefore up to us, as schools and parents, to ensure technology serves pupils and their learning, rather than becoming an end in itself.
DIGITAL LEARNING: WHY ISN’T IT WORKING?
01
READING ON SCREEN: THE SCREEN-INFERIORITY EFFECT
Reading on screens triggers an unconscious shift from deep comprehension to shallow skimming.
Research shows that when we read on paper we comprehend and remember the text better than when we read it on a screen. This is partly due to cognitive mapping, where the brain ties new information to a specific location. When we scroll through text on a screen, the words have no fixed location making it harder for the brain to anchor and retrieve memories of the text.
Recent research confirms the screen-inferiority effect:
- This meta-analysis of 171,055 participants across 54 studies found that paper-based reading produced significantly better comprehension outcomes than digital reading. The advantage of paper was greater when readers were under time constraints.
- Jerrim et al 2018 looked at whether taking the PISA test digitally or on paper had an impact on performance. The study found the paper group scored roughly 20 points higher on PISA tests – the equivalent of six months additional learning.
02
WRITING BY HAND: THE LONGHAND SUPERIORITY EFFECT
Research shows that taking notes by hand leads to significantly better learning than typing notes on a computer.
Handwriting activates a broader network of brain regions involved in motor, sensory, and cognitive processing. Typing engages fewer neural circuits, and results in more passive cognitive engagement.
Writing notes by hand is slower than typing and this is its strength. Pupils have to identify what matters most, summarise and reorganise information as they write – all of which encourages deeper cognitive engagement.
Thus handwriting activates thinking in ways that typing never will and displacing handwritten notes with digital alternatives deprives children of a critical tool for learning and memory retention.
03
ATTENTION AND THE IMPACT OF TASK-SWITCHING
Multitasking is not a thing. Multitasking is fragmented attention switching.
Elizabeth Stone – Head Teacher, Winchester College.
Every child knows what an exercise book is for in the classroom – learning. Screens are for…..everything. They are a tool for infinite tasks. Scrolling, music, doing homework, watching movies, checking your calendar, shopping, gaming, messaging friends. So when a child sees a screen, their brain goes into task-switching mode.
When humans switch tasks, three things happen:
- Time is lost: between tasks there is a brief moment when the attention shuts down. Although each “mental blink” only lasts around 0.15 seconds, effects on attention are cumulative.
- Accuracy falls: Each time the brain swaps mode – responding to notifications, seeing the internet explorer icon and being reminded of those trainers you were looking at last night, an email pinging – the competing modes bleed together making it much more likely we’ll commit an error.
- Memory reduces: When we’re focused, the brain uses the hippocampus to build strong, lasting memories. When we shift tasks, it uses a different region of the brain – the striatum – that’s specialized for automatic routines and habit rather than deep learning. When information is stored as a habit it is much harder to actively recall and apply later.
Studies also show that pupils are more likely to work off-task when using devices in the classroom. One study revealed that students spend approximately 38 minutes of every hour off-task while using a laptop for educational purposes.
04
WHAT ABOUT AI?
Research published since 2024 increasingly demonstrates that AI can undermine learning. Studies point to AI creating the illusion of learning while leaving pupils with less knowledge and understanding than those who learn without it. For example, in June 2026, research by Strömberg and colleagues followed more than 26,000 secondary students over the course of two and a half years. Those who began using AI for their homework completed it 30% faster than before and their marks rose by 18%, but in exams taken without AI, their scores fell by up to 24%. Even pupils who preferred using a chatbot over taking notes, finding it more helpful, understood and remembered significantly less (Kreijkes 2026). Generative AI also doesn’t treat pupils fairly. It discriminates against pupils’ sex, ethnicity, SEND and EAL status using cues from the input pupils give it (Tan, 2026).
Since August 2026, Gemini has been switched on by default in Google Classroom for UK pupils of all ages, AI summaries in Google searches cannot be switched off and AI mode is embedded in help features in many tools.
Almost two-thirds of parents worry their children are using AI for their schoolwork while two-thirds of secondary teachers warn that pupils’ critical thinking has already declined due to the use of Generative AI in particular. As yet no large, independent trial has shown lasting learning gains for children over the same teaching without AI. With no demonstrated lasting benefit to set against these risks, it is wise to adopt the precautionary principle until AI applications are proven to be safe and not damaging to cognition.
05
THE HEALTH IMPACTS OF INCREASED SCREEN TIME.
We worry about school meals and sport, and yet we throw caution to the wind with screens, even though we now know they adversely impact children’s development and their learning.
Becky Foljambe, GP and Founder of Health Professionals for Safer Screens
In new guidance for schools on technology use in the school day, Health Professionals for Safer Screens warn:
Rising Myopia (Short-sightedness): 20% of children in the UK have myopia, up from less than 10% 50 years ago. Every additional hour per day of digital screen time has been associated with a 21% increase in the chances of developing myopia, which carries a life-long raised risk of retinal detachment, glaucoma and cataract.
Cardiometabolic Health: Screentime exceeding 2 hours per day is consistently associated with an increased risk of being overweight or obese; metabolic syndrome; and hypertension. When screens or 1-1 devices displace physical and social activity at school, many children are deprived of their only opportunity to move, play and socialize in the real world.
Blue light and sleep: A National Sleep Foundation consensus panel achieved formal consensus that screen use causally impairs sleep health. Device use, especially in the evening, is consistently linked to poorer sleep and daytime sleepiness. When homework is set on a screen, parents are left with a difficult choice: allow the homework and disrupt sleep, or prioritise sleep and disrupt homework.
ADHD and attention-related symptoms: Screentime is consistently associated with both ADHD symptom severity and autism risk, with longitudinal evidence confirming digital media use predicts later ADHD symptoms – not merely the reverse.
WHAT CAN WE DO ABOUT IT?
Parents and teachers need confidence that every tool in the classroom is safe, effective and genuinely improving outcomes.
Paul Waugh, Parliamentary Under-Secretary of State for Early Education.
At Generation Focus we’re developing a set of resources to help schools review their use of technology. Our education team works on the ground with schools who are reviewing their digital strategies at primary and secondary level.
This toolkit includes five practical steps schools can take to make immediate changes in light of the new evidence, while working towards longer-term changes that require more consideration and preparation. It draws on the latest research and expert opinions and expert opinion from educationalists, psychologists, researchers and school leaders.
This is a list of FAQs about reducing AI & EdTech in schools we hear from school leaders and parents, and evidence-based answers.
If you’d like to speak to us, please get in touch at hello@generationfocus.org and we’d be delighted to support you.
Sign up to our newsletter here.
AI and EdTech

