Xina Crea Dos Policies Virtuals A Internet
Linda Berens
The majority of national AI strategies recognise the value of adopting AI in the public sector, alongside the need to mitigate its risks (OECD/CAF, 2022; OECD 2019). In fact, governments are increasingly using AI for public sector innovation and transformation, redefining how they design and deliver policies and services. While the potential benefits of AI in the public sector are significant, attaining them is not an easy task. The field is complex and has a steep learning curve, and the purpose and context of government presents unique challenges. In addition, as in other sectors, public sector algorithms and the data that underpin them are vulnerable to bias, which may cause harm, and often lack transparency.
Governments are also increasingly recognising the specificities of gig economy workers and making efforts to integrate their perspectives in policy making. For instance, in Seattle (United States), the city used human-centred design to engage with drivers and gain a deeper understanding of their preferences with respect to potential policies on minimum compensation. To this end, the local government developed an engagement strategy including elements of ethnographic analysis that resulted in interviews, focus groups, a telephone town hall and an online survey. Listening to the voices of gig economy drivers enabled the city of Seattle to ensure that innovation efforts would actually address their needs, as evidenced by a recent OECD Innovation and Data Use in Cities report.
Trend 1 discussed the importance of algorithmic accountability in the public sector as a means to ensure that AI systems increasingly playing a role in government decision making are transparent and fair. However, what if in addition the public could play a role in elaborating the policies that apply these algorithms in their community? The WeBuildAI participatory framework helps to show governments how such an approach could take shape (Box 46).
aa06259810