Perceptions and assumptions about the nature of disability and its causes influence AI designers and shape the design of AI products. Disability Studies is a multidisciplinary academic field concerned with understanding disability and improving the lives of Persons with Disabilities (PWD). Disability frameworks developed within this field have been integrated to improve design and policies in many sectors, including mobility, healthcare and welfare. In the last few years, several activists and scholars have used these disability frameworks to speak out against the ableist perceptions and practices dominating the AI research and industry. However, very few efforts have been made to leverage these frameworks to improve the design of AI systems in the context of disability. This paper is an attempt to mind and bridge this gap. We will first focus on the concept of “Disability Creation” in the digital context and, more specifically, AI. We will examine conceptual, ethical, and legal disability frameworks and identify factors that contribute to the creation or alleviation of disabling situations linked to the use of AI. Finally, we will map these factors in a novel design-oriented framework for inclusive AI called D-DCP and inspired by HDM-DCP, the “Human Development Model and Disability Creation Process”.

The rise of news content on social media has been accompanied by a hope that people with lower socioeconomic status and less interest in political affairs would be “accidentally” exposed to news. By combining tracking and survey data from a Dutch online panel (N = 413), we analyze how political interest, income, and education influence social media news exposure and consumption. Higher levels of political interest are associated with higher amounts of news exposure on Facebook and more news items consumed via social media. Users engage less often in news-related follow-up behavior after consuming news items via social media than after consuming news items referred via news websites. If social

News recommenders help users to find relevant online content and have the potential to fulfilla crucial role in a democratic society, directing the scarce attention of citizens towards the information that is most important to them. Simultaneously, recent concerns about so-called filter bubbles, misinformation and selective exposure are symptomatic of the disruptive potential of these digital news recommenders. Recommender systems can make or break filter bubbles, and as such can be instrumental in creating either a more closed or a more open internet. Current approaches to evaluating recommender systems are often focused on measuring an increase in user clicks and short-term engagement, rather than measuring the user's longer term interest in diverse and important information. This paper aims to bridge the gap between normative notions of diversity, rooted in democratic theory, and quantitative metrics necessary for evaluating the recommender system. We propose a set ofmetrics grounded in social science interpretations of diversity and suggest ways for practical implementations.