While it’s perhaps fair to say that governments have not been at the front of the queue in terms of adopting AI, that should not mean that the potential is not considerable.  A new paper from the University of Oxford aims to provide some practical lessons to help guide government AI projects.

“Governments around the world are launching projects that embed AI in the delivery of public services.  These range from AI-driven management of internal systems to smart city solutions for urban problems,” the authors say.  “Yet many of these projects fail due to lack of financial resources, poor oversight, or knowledge gaps.  We believe there is a clear need for a succinct framework that will help government decision-making navigate the complexities of AI projects, avoid pitfalls and uphold the public good.”

Best practice advice

The paper explores a number of AI projects currently in scope to try and uncover any best practice advice.  These projects were grouped into four distinct types:

• Reformer projects – which tend to have high resources and high project importance.
• Steward projects – which tend to have high resources but relatively low importance.
• Aspirant projects – which tend to have high importance but limited resources.
• Adventurer projects – which tend to have both low resources and importance.

This classification system was then used to develop five practical principles to help governments manage the varied and diverse range of AI projects they might be engaged in, while also minimizing any risks to the public:

• Determine appropriate solutions – decision-makers should critically assess whether and how AI can help governance challenges
• Include a multi-step assessment process – consider using feasibility studies, pilots, milestones for quality control, and post-implementation monitoring
• Strengthen government’s bargaining position – robustly engage with technology vendors and external partners using tactics such as blacklists and bulk tenders
• Pay attention to sustainability – ensure human talent is available as well as financial and political support for long-term success
• Manage data, cybersecurity, and confidentiality effectively – protect the national interest and individual privacy, as well as win public trust

“In our study, we have shown how certain practical principles of good governance can be deployed to mitigate the risk or pursue advantages inherent in different types of AI projects,” the authors conclude. “By following this approach, we hope that government officials will benefit from a greater awareness of the risks, opportunities, and strategies suitable for their particular project. “