Public Procurement as AI Governance Tool

During my PhD projects, I found that although there are many regulations, standards and guidelines for digital technology, in practice there is still a serious lack of transparency and accountability. The organisations responsible for oversight often lack the tools and resources to monitor the sector effectively, which makes enforcement difficult. This led us to explore alternative ways of enforcing legal principles, rights and obligations in the digital space.

With my past experience in sustainable procurement, I began to think about the role the public sector could play in shaping how AI is designed, developed and deployed. After all, the public sector is the largest customer of technology in terms of spending. What if it started applying stricter checks at the gate?

Together with Emma Kallina, I am exploring how public procurement can be utilised as an AI governance tool. Public procurement refers to the process by which public and semi-public bodies purchase products or services. It gives them the power to set conditions for who they buy from and the standards they expect.

Recently, there has been growing interest in how the public procurement of AI could be strengthened. To better understand this, we spoke with experts, buyers and suppliers involved in public sector AI procurement. Our aim was to identify which concrete levers are actually available to help ensure that the use of AI in the public sector becomes more responsible, and what should be considered in the implementation of these.