Asker Municipality and the University of Oslo tested subject-adapted conversational AI chatbots in classrooms (~200 pupils, 2024; three lower-secondary schools, 2024/25); researchers found most pupils need explicit teacher modelling to use the tools well.
~200
Pupils observed in spring 2024 phase (spring 2024)
5
Primary schools in spring 2024 phase (spring 2024)
7
Classes observed in spring 2024 phase (spring 2024)
3
Lower-secondary schools in 2024/25 phase (autumn 2024-winter 2025)
Details
Maturity
Scaling
Promoter
Asker Municipality with the University of Oslo (Institutt for lærerutdanning og skoleforskning)
Period
2023–2026 (three-year project)
Keywords
K-12 education, conversational AI, teacher professional development, learning research
Context
«Læring i algoritmenes tidsalder» (LAT — “Learning in the Age of Algorithms”) is a three-year innovation project funded by the Research Council of Norway, with Asker Municipality as project owner and the University of Oslo's Institutt for lærerutdanning og skoleforskning as research partner.
Activities
In spring 2024, LAT researchers visited five primary schools and observed six teachers using age- and subject-adapted conversational chatbots with just under 200 pupils across seven classes. In autumn 2024 and winter 2025, researchers observed chatbot use in social studies and science over two-to-six-week periods at three lower-secondary schools.
Results
The research found that lessons using conversational chatbots are demanding for most pupils, both in content and dialogue, and that most pupils need explicit teacher modelling and close follow-up to use the tools productively — a finding that fed into municipal guidance rather than being presented as unqualified success. Asker has since built its own set of six subject-specific chatbots (for concepts, writing, language, reading and coding), inspired in part by neighbouring Randaberg municipality's earlier work.
Implementation
Indicative cost
Medium (€50k–€500k) — Funded as a Research Council of Norway innovation project; no public total budget figure found.
Time to results
Medium (1–3 years) — Three-year project (2023-2026); observation phases in spring 2024 and autumn 2024-winter 2025.
Staffing & skills
University of Oslo (ILS) researchers, Asker Municipality digitalisation team, Classroom teachers piloting chatbots
Conditions for success
Explicit teacher modelling and close follow-up, since most pupils need this to use conversational AI productively
Municipal capacity to build and maintain subject-specific chatbots after the research phase
Common failure modes
Lessons using conversational chatbots proved demanding for most pupils in both content and dialogue when teacher modelling was absent
Commonly funded by
National / regional programmes
Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.
Do you run this practice?
Claim it —
verified implementers get a public contact pathway and can propose corrections.
Data sources
Where this practice's information was retrieved from, and when.
MBRGI's Digital School launched a three-year AI-personalised learning rollout across six African countries including Mauritania, pairing 100 digital learning spaces …