evidoria

← Back to browse

Good practice Imported

Gemini for Education at Italy's Salesian (Don Bosco) Schools

Italy · Verona · See the Italy profile

Evidence: Observational / pre–post Top 59% 47/100 · Ask Evidence Copilot about this practice

Italy's Salesian school network scaled a Gemini for Education pilot from 700 Triveneto teachers to 1,600+ nationally in a year; a Pontifical Salesian University study of 29,000+ students found +9% engagement, +16% creativity, and 69% of teachers reporting improved lesson planning

1600+ teachers
Teachers using Gemini for Education (by May 2026)
29000+ students
Students reached (by May 2026)
9 %
Rise in student engagement (Pontifical Salesian University analysis)
16 %
Rise in student creativity (Pontifical Salesian University analysis)
21 %
Rise in teacher engagement (Pontifical Salesian University analysis)
40 %
Rise in teacher creativity (Pontifical Salesian University analysis)
69 %
Teachers reporting improved lesson planning (teacher survey)
86 %
Teachers wanting to continue using generative AI (teacher survey)
70 %
Reduction in teacher administrative time (reported separately)
Gemini for Education at Italy's Salesian (Don Bosco) Schools

Details

Maturity
Scaling
Promoter
Google for Education; Congregazione Salesiana / Istituto Salesiano San Zeno; Università Pontificia Salesiana (research)
Period
pilot from Triveneto (~2025) to national rollout by May 2026
Region (NUTS)
ITH31
Keywords
K-12 & vocational education, generative AI, teacher lesson-planning, EdTech

Context

Starting in Italy's Triveneto region and expanding nationally within about a year, the Salesian (Don Bosco) school network - coordinated with Istituto Salesiano San Zeno in Verona - partnered with Google for Education to deploy Gemini for lesson planning, content personalisation and administrative tasks.

Activities

The programme grew from 700 teachers to more than 1,600 teachers and over 29,000 students by May 2026. A Pontifical Salesian University-coordinated analysis, led by researcher Don Michal Vojtáš, examined 1,375 classroom activities across the network.

Results

The analysis found a 9% rise in student engagement and a 16% rise in student creativity, alongside a 21% rise in teacher engagement and a 40% rise in teacher creativity. In teacher surveys, 69% reported improved lesson planning and 86% said they wanted to continue using generative AI tools. A separate account of the same programme reported a 70% reduction in teachers' administrative time, reallocated to one-to-one student mentorship.

Conclusions

The school network paired the rollout with an explicit pedagogical boundary - 'when the bell rings, Gemini for Education stops' - confining AI use to preparation and admin rather than classroom relationship-building. The findings come from teacher self-report surveys and activity-log analysis, not an independent controlled trial.

Implementation
Indicative cost
Low (< €50k) — No specific budget figure is reported; the programme runs on Google for Education's Gemini tools within the existing Salesian school network.
Time to results
Short (< 1 year) — Piloted in Triveneto (~2025), reaching national scale (1,600+ teachers) within about a year, by May 2026.
Staffing & skills
Google for Education provided the Gemini platform/tools, Congregazione Salesiana / Istituto Salesiano San Zeno (Verona) coordinated the school-network rollout, Università Pontificia Salesiana (Don Michal Vojtáš) conducted the research analysis

Conditions for success

  • Explicit pedagogical boundary confining AI use to preparation/admin, not classroom relationship-building ('when the bell rings, Gemini for Education stops')
  • Phased rollout starting in one region (Triveneto) before national expansion
  • Partnership between an existing school network, a technology vendor (Google) and an academic research partner for evaluation

Common failure modes

  • Findings are based on teacher self-report surveys and activity-log analysis, not an independent controlled trial
  • No equity/inclusion-specific targeting or breakdown is reported for the rollout

Where it fits

Governance type
faith-based private school network in partnership with a technology vendor and a university research team
Scale
national (1,600+ teachers, 29,000+ students) grown from a single-region pilot
Income level
high-income

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