OxCAIGG: AI in the Public Service: From Principles to Practice
13 December 2021
How do people perceive the risks of using AI (Artificial Intelligence) and machine learning tools in decision making? Using survey data from a sample of 154,195 respondents in 142 countries collected for the 2019 World Risk Poll, we analyze basic indicators of public perceptions about the potential harms and opportunities of involving AI in our personal affairs and public life. Public understanding of AI—its risks and benefits—varies greatly around the world.
Our findings suggest that putting AI to work for good governance will be a two-fold challenge. Involving AI and machine learning systems in public administration is going to require inclusive design, informed procurement, purposeful implementation and persistent accountability. Additionally, it will require convincing citizens in many countries around the world that the benefits of using AI in public agencies outweighs the risks.
Cite as: Written & researched by Lisa-Maria Neudert, Aleksi Knuutila, Philip N. Howard (2020). Global Attitudes Towards AI, Machine Learning & Automated Decision Making. Working paper 2020.10 Oxford, UK: Oxford Commission on AI & Good Governance, 10pp.