evidoria

← Back to browse

Good practice

FAIE — Fair Artificial Intelligence Educator: AI-Adaptive Maths Practice for Maltese Primary Schools

Malta · Msida · See the Malta profile

A University of Malta-built AI app adapts maths practice for pupils aged 8–10; a small controlled pilot of 280 students found low-performing students in the AI group gained 23% versus no significant gain in the control group — real but developer-reported results.

23 %
Average improvement for low-performing students in FAIE (experimental) group (pre- to post-test, pilot period)
72 %
Students preferring FAIE over paper worksheets (pilot period)
280 students
Students in controlled pilot (control + experimental groups) (pilot period)

Details

Maturity
Pilot
Promoter
University of Malta, Faculty of ICT (Prof. Alexiei Dingli et al.)
Period
Concept 2019 (Malta Social Impact Award); controlled pilot reported in 2020 (ICERI2020) and 2023 (INTED2023) conference proceedings
Keywords
primary education, mathematics, adaptive learning, low-performing student support, academic research

Context

FAIE (Fair Artificial Intelligence Educator) is an AI-adaptive mathematics practice app for Maltese primary-school pupils (Years 4-6, ages 8-10), created by Prof. Alexiei Dingli and colleagues at the University of Malta's Faculty of ICT. The concept won Malta's Social Impact Awards in November 2019, receiving €50,000 in prize funding plus pro-bono mentoring support.

Activities

The developers ran a controlled pilot with 280 students split into control and experimental groups, reported in conference proceedings at ICERI2020 and INTED2023.

Results

Low-performing students using FAIE showed a 23% average improvement between pre- and post-tests, while the control group showed no significant change. 72% of surveyed students said they preferred the app to paper worksheets, and teachers found it easy to use.

Conclusions

These results come directly from the tool's own academic creators rather than an independent evaluator, and the published papers do not specify the exact number of participating schools. No source confirms whether FAIE has been used beyond this original pilot or what its current deployment scale is — this should be read as evidence of a promising, small-scale, developer-evaluated trial rather than a proven, at-scale intervention.

Implementation
Indicative cost
Low (< €50k) — €50,000 in prize funding from Malta's Social Impact Awards (November 2019), plus pro-bono mentoring support; no further public budget figures for development or piloting costs are available.
Time to results
Short (< 1 year) — Award and initial development in 2019; pilot results published in conference proceedings at ICERI2020 and INTED2023; no confirmed activity or evaluation after that period.
Staffing & skills
Prof. Alexiei Dingli and colleagues, University of Malta Faculty of ICT (developers and evaluators)

Conditions for success

  • Direct academic ownership by the app's developers enabling close integration between the pedagogical model and the technology
  • Recognition and seed funding via Malta's Social Impact Awards (2019), including pro-bono mentoring support
  • Targeting a specific, well-defined at-risk group (low-performing students in Years 4-6) rather than a general population

Common failure modes

  • Evaluation was conducted and reported by the tool's own creators, not an independent evaluator, so findings should be treated with more caution than an independently-audited study
  • Published papers do not specify the number of participating schools, limiting assessment of how representative or generalisable the pilot sample was
  • No evidence exists of use beyond the original pilot or of current (2026) deployment scale

Where it fits

Governance type
university-led academic pilot, not a government-mandated programme
Scale
single small-scale pilot (280 students)
Income level
high-income

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful