Ethics of Using Smart City AI and Big Data: The Case of Four Large European Cities

Abstract

By 2030, the population living in cities will increase by an additional 1.5 billion people, placing a great strain on resources, infrastructure, jobs and healthcare (UN 2018). It has become clear that to combat this change, a number of creative approaches need to be put in place to ensure the sustainable growth of cities - one such approach is the ‘smart city’ (UN 2018). Due to the relative infancy of smart cities, and the diversity of approaches and implementations of smart information systems (Big Data and AI), many of the ethical challenges are still being defined.

One of the reasons behind this challenge is a result of the varying smart information systems (SIS) being used in different urban contexts. This case study aspires to unpack some of these ethical challenges by looking at four different applications of SIS being deployed in large European cities: an AI used to understand citizens’ complaints (Amsterdam), a parking permit chat-bot (Helsinki), a platform for data exchange (Copenhagen), and a project with an open-source algorithm (Hamburg).Upon first glance, these technologies seem very disparate, but they all factor into the equation of what goes into making a smart city, ‘smart’.

Over the course of the interviews, what quickly became clear was the degree to which smart cities are in their infancy, meaning that the availability and accuracy of data remains an issue in a large majority of the cases. In terms of the accuracy of recommendations – due to the early stages of smart city implementation, many projects remain wary of expanding the use of SIS, due to potential unforeseen issues and are therefore proceeding cautiously.

Data has been taken on as a potentially helpful tool for citizens and planners alike to regain control and access to information within their respective cities. Consent, transparency and data ownership featured as prominent ethical considerations in all cases, especially the focus on citizens regaining control over their own data. Further, it remained a point of contention to whom the data would belong – with an overall consensus that data should remain the property of the citizen or municipality and not necessarily that of private companies.

Throughout the process, it became clear that collaboration is at the heart of a successful smart city. Many of the projects utilised a collaborative public-private model to facilitate both the business development side and the citizen-engagement sides of the smart city. With differing degrees of success in the individual projects, this remained an important feature that experts believe will continue to develop in tandem with smart city projects. A bottom-up approach is clearly the most effective way to ensure that a smart city works and is used by its citizens.

Overall, this case study offers valuable insights into the development of smart cities in a European context: including the use and implementation of SIS in urban environments, what kinds of ethical issues are evaluated in the literature and how they contrast and diverge from those faced by professionals in practice. It is hoped that this case study will offer practitioners, policymakers, smart city organisations, and private ICT companies interesting observations about a more ethically responsible approach towards SIS implementation in smart city projects.