evidoria

← Back to browse

Good practice Imported

Osloskolen's Own School Chatbot — Oslo Municipality's Privacy-Compliant Alternative to Commercial AI, with a 236-Teacher Early-Adoption Study

Norway · Oslo · See the Norway profile · See the Oslo profile

Top 70% 40/100 · Ask Evidence Copilot about this practice

In October 2023 Oslo launched its own school chatbot as a privacy-compliant alternative to commercial AI. A peer-reviewed survey of 236 teachers found self-efficacy drove perceived usefulness while management support was weak; learning outcomes remain unmeasured.

Osloskolen's Own School Chatbot — Oslo Municipality's Privacy-Compliant Alternative to Commercial AI, with a 236-Teacher Early-Adoption Study

Details

Promoter
Oslo Municipality (Osloskolen / Education Agency)
Period
2023-present
Keywords
school chatbot, generative AI, data protection, teacher adoption, lower and upper secondary

Description

In October 2023 Oslo Municipality launched a school-specific chatbot so that teachers and pupils in lower and upper secondary school could use generative AI through a municipal service built to comply with privacy requirements, rather than through commercial tools. The municipality states that the service does not store personal data and that it gives all Osloskolen pupils equal access to explore AI. Staff guidance asks teachers to check AI output critically, cite AI-generated material, have a clear purpose for using AI, give pupils clear guidance, and never let AI replace professional judgement; pupils have a separate guide on responsible use.
Elstad and Eriksen (Nordic Journal of Comparative and International Education, 2024) surveyed upper-secondary teachers six to eight weeks after launch (N = 236, response rate about 96%) and used structural equation modelling to study what predicted perceived instructional usefulness of the chatbot. Teachers' instructional self-efficacy had the strongest positive link to perceived usefulness. Links from school management and from colleague discussion were weak, which the authors read as untapped potential, noting that no national or local guidelines for AI in schools existed when the survey was run. The launch was deliberately low-key and was discussed with the teachers' unions in December 2023.
Cautions: the evidence is a single early-stage perception survey from one city; it measures teacher attitudes, not pupil learning, and the authors note findings may change as teachers gain experience. No learning-outcome evaluation was found. The value of the practice is as a documented governance model (municipal, no-personal-data-storage chatbot) and as a signal that teacher capability and management support, not just tool access, determine uptake.

Read the full analysis: https://www.oslo.kommune.no/skole-og-utdanning/digitale-verktoy-osloskolen/kunstig-intelligens-ki-i-osloskolen/

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

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.

Attachments

Similar practices you may find useful