Report published: 6 October 2026
Scope: England
Reviewed: 12 October 2026
A new independent randomised trial has examined Oak National Academy’s free AI lesson-planning assistant, Aila.
The Education Endowment Foundation commissioned NFER to evaluate the tool with Key Stage 2 teachers over ten weeks in autumn 2025. The final analysis included 416 teachers from 108 primary schools. Teachers encouraged to use Aila spent about two hours and 30 minutes a week preparing lessons and resources, compared with three hours and 19 minutes in the comparison group: an average saving of 49 minutes, or 24%. Independently assessed resources were no worse in quality.
Adoption was uneven. Eighty-one per cent used Aila at least once, but it appeared in only around 7% of lessons during an average week and use declined over the trial. Fewer than one in five teachers said it meaningfully changed their teaching or planning.
The comparison group could use other generative-AI tools, and the trial did not measure pupil learning. The findings therefore support a limited conclusion: a specialist tool can reduce planning time for some teachers, but free access alone does not ensure sustained use or better teaching. Professional checking and adaptation remain necessary.
Primary sources: EEF project and evaluation and EEF findings summary.