Towards a Systematic Understanding on the Challenges of Procuring Artificial Intelligence in the Public Sector
By Keegan McBride,, Colin van Noordt, Gianluca Misuraca, and Gerhard Hammerschmid
There is an increased interest amongst governments and public sector organizations about how to best integrate artificial intelligence into their day-to-day business processes. Yet, a large majority of technical knowhow is concentrated within the private sector, requiring most public sector organizations to rely on public procurement for AI systems. While many governments may have experience with traditional forms of public technological procurement, this paper argues that the public procurement of AI is different and new insight is needed to understand this differentiation and procure AI better. This paper offers an initial contribution to the public administration and management literature by describing this difference, and identifying the challenges associated with the public procurement of AI. In order to achieve this contribution, the research studied guidelines in four European countries (Estonia, Netherlands, Serbia, and the United Kingdom) to generate insight into the challenges faced, and potential solutions to these challenges, during the public procurement of AI process.