[Return to ELISE's Artificial Intelligence in the public sector page]
This report is published in the context of AI Watch, the European Commission knowledge service to monitor the development, uptake, and impact of Artificial Intelligence (AI) for Europe, launched in December 2018. As part of the AI Watch initiative, a specific task addresses the role of AI in the public sector, which sets out to provide actionable guidelines to promote the adoption of safe, lawful, inclusive, and trustworthy AI by public sector administrations in the EU.
This report contributes to the existing knowledge on Artificial Intelligence in the Public Sector, by moving from a more theoretical and anecdotal view to more systematic analysis. It is based on analysing a collection of original evidence about how AI is being used in the public sector, its impacts, challenges, and barriers.
The report is built on three pillars of research:
- the National Strategies published by Member States on AI, with a specific focus on how those strategies address public sector related challenges to AI development and use;
- an inventory of AI use cases in the public sector; and
- an in-depth view of 8 cases of AI use in the public sector, where a variety of “on-field” information has been collected through case study research.
The mapping of all 27 EU Member States- plus Norway- national strategies is an update of the previous work performed by Misuraca & van Noordt (2020). It consisted of a deep dive into each strategy and a comparison among the policy initiatives to stimulate the development and adoption of AI in the public sector. The report also presents a factsheet per country (Annex 1) synthesizing the main elements of their AI strategy.
The inventory consists of 686 AI use cases implemented in all EU Member States and some other countries in Europe. This collection aims to be the first attempt to create a repository of AI cases at the European level, to investigate risks, and provide researchers and policymakers with a database of cases for fostering research and exchange. A taxonomy system was designed to describe features and provides insight into the technological dimensions of AI, techniques applied, administrative context, and its relation with the OECD's Framework. The cases selected for an in-depth analysis can support and inspire other public organisations in implementing AI. The database is available as open data in the JRC open data catalogue. Explore Annex II for detailed description of the case studies.
The catalogue provides evidence that AI has moved beyond rare experiments and pilots to a rich set of concrete applications and solutions applied daily to increase government efficiency and effectiveness
The report highlights that the diffusion of AI remains unequal, and the breaking down of barriers to AI adoption requires significant consideration by policymakers. Progress in this area will depend on ensuring the right balance between public and private sector expertise and capacity, strong collaboration, enhanced data governance, and risk mitigation.
The report provides policy recommendations at the organisational level that should be considered complementary to those published in the report “Road to the adoption of AI by the public sector” (Manzoni et al., 2022).
Explore the Annexes of the Report:
How to cite this report: Tangi, L., Van Noordt, C., Combetto, M., Gattwinkel, D. and Pignatelli, F., AI Watch. European landscape on the use of Artificial Intelligence by the Public Sector, EUR 31088 EN, Publications Office of the European Union, Luxembourg, 2022, ISBN 978-92-76-53058-9, doi:10.2760/39336, JRC129301.