Governance and Human Rights
Sandbox for responsible AI public procurement
Objectives
The project aims to pilot a procurement sandbox to help GPAI member states assess AI products and services they have procured, helping to test and improve compliance with responsible AI procurement guidelines or regulations, and identify possible gaps and challenges.
The project seeks to co-develop and test procurement sandbox prototypes with GPAI governments for specific use cases.
We plan to engage with governments in co-creation workshops to identify use cases of interest based on their needs and aspirations. Then, we will develop, test and adapt the prototype sandbox in an iterative design process. The project outcomes include the procurement sandbox functional specification and testing protocol, as well as lessons learned from initial testing and a template for implementing the sandboxes based on the tested mockup specification.
The project extends current efforts for operationalizing AI governance, and can lead to more agile procurement guidelines, as well as benchmarking, standards and metrics. It will also serve as further evidence for the benefits and challenges of using sandboxes in different sectors and environments.
Team
Lama Saouma
AI Initiatives Lead
Juan David Guetierrez
GPAI Expert and project co-lead
Arunima Sarkar
GPAI Expert and project co-lead
Steering Committee
Aditya Mohan
Tom Lenaerts
Rob Heyman
Virginia Dignum
Juliana Sakai
Clara Neppel
For this project, we are currently engaging with the World Economic Forum and might work with different governments through WEF’s Centres for the Fourth Industrial Revolution in multiple countries.
Governments who would like to pilot a sandbox for AI procurements are welcome to reach out to us so that we can identify the use cases of interests together.
Additionally, we welcome expert representatives from external organisations in this domain to join the team as members of the project advisory Group.
