Catalogue: AI in the Public Sector clear
The UK i.AI built Redbox, an open-source RAG chatbot enabling civil servants to interrogate policy documents via LLM. Deployed to 6,000+ staff …
Ecuador's SRI deploys Falcon, cross-referencing 400+ data types to assign each taxpayer a 0–1,000 risk score. In 2023 it identified 60,000 previously …
BPS (Statistics Indonesia) deployed generative AI to auto-classify 42 million business entities against the KBLI 2025 standard during the 2026 Economic Census, …
Bangladesh's NISE (2021) uses AI job-matching and skills forecasting to connect youth to employment. Over 1M registered, 300K+ job opportunities created, 2,000+ …
Saudi Ministry of Interior and SDAIA deploy Baseer (tracking 1M+ pilgrims/day via computer vision) and the biometric Makkah Route Initiative across 11 …
Rwanda's Ministry of Health built an AI Health Operating System connecting EMRs, DHIS2 and logistics data. The Health Intelligence Centre (launched April …
Since 2014 Denmark’s Skattestyrelsen has run a real-time ML system flagging suspicious VAT refund claims; it reports ~70% accuracy, is now integrated …
Portugal's AMA launched an AI chatbot on ePortugal.gov.pt in May 2023. By December 2023, 23,780 conversations (101/day) and a 10% rise in …
From 2022, Australia's ATO deployed ML to detect a large-scale GST refund fraud scheme: 57,000+ alleged offenders identified, A$2.7 billion in additional …
Seven-month pilot (from May 2025) in which three German authorities simulated an EU AI Act regulatory sandbox with two medical-device companies, ahead …
The Swiss Federal Supreme Court trained its own judgment-anonymisation model (92.26% recall, 92.57% precision, self-reported) on two in-house GPUs and runs ChatTF …
Antibes Juan-les-Pins has published an inventory of 8 municipal algorithmic processes since 2021, updated every six months under France's Lemaire law; only …
After a 2020 QPC by the UNEF student union, France's Constitutional Council ruled that universities must publish, once admissions end, the criteria …
Helse Bergen tested an AI early-warning tool for patients at high risk of readmission inside Norway's DPA sandbox. The exit report (Sept …
Launched by San José in 2023, the GovAI Coalition shares an AI FactSheet, model AI policy and incident-response templates among roughly 150 …
Ontario approved 20 AI scribe vendors for physicians (5,000+ participating). The Auditor General's May 2026 testing found 9 fabricated content, 12 swapped …
Brazil's electoral court runs a public Sistema de Alerta de Desinformação Eleitoral with CIEDDE, a direct channel to platforms. Reports fell from …
Austria's regulator RTR runs an AI Service Desk with an advisory board, events and an open survey of 700+ firms on AI …
Brazil's electoral court runs an official WhatsApp bot that searches fact-checks by Lupa, Aos Fatos, Comprova and Estadão Verifica. It reported 530,000+ …
UNESCO and Digitaal Vlaanderen published (2 June 2026) the first regional AI Readiness Assessment, piloting RAM 2.0; it maps governance gaps but …
Regulation 10, proposed April 2023 and in force from 7 Sept 2023, sets notice, audit and officer duties for AI processing personal …
Ontario's AI Directive (in force 1 Dec 2024) requires ministries and agencies to manage AI risk and publish use cases; regulators say …
Singapore's new Institute of Digital Government ran pilot 'AI Sprints' — agency workshops applying AI to real work — reaching 1,000+ officers …
Nigeria's central bank added a machine-learning sentiment index to its inflation forecasting model, improving accuracy; an independent 2025 university study on the …
Open full copilot Grounded in cited practices — always check the sources.
