Anniek Jansen

In the Netherlands, several public health platforms provide information and peer support to different patient groups such as people suffering from cancer or multiple sclerosis. With the rapid growth and development of Large Language Models (LLMs) in multiple sectors, these health platforms are considering the use of LLMs for personalized content generation and ease of information management. However, the sensitive nature of health information and personal experiences shared on such platforms introduces privacy and ethical risks, such as misinformation and bias. This study explores the concerns and risks associated with LLMs through interviews and focus groups with platform editors and users who use and access these platforms. Our findings show that risks related to content quality and quantity were most frequently identified. Moreover, the findings highlight the importance of disclosure if there is no human oversight, participants’ strong opposition to AI-generated blogs, and the potential of LLMs for personalization, provided users retain control over what they read. This work contributes to the ongoing discussion in human-centered computing about the ethical challenges and risks of adopting LLMs by presenting an empirical evaluation with editors and users. Moreover, the insights inform considerations and design guidelines for implementing LLMs on public health platforms.

The surge of Artificial Intelligence (AI) for automatic decision-making raises concerns about transparency and interpretability of AI models. Explainable AI (XAI) addresses this by providing insights into AI predictions. Despite the availability of various methods for explaining decisions based on tabular data, there is no consensus on their effectiveness for different types of users. This paper introduces a novel XAI method, the Visual Map, and presents a human-grounded evaluation study comparing it with three common XAI methods. In an online experiment (N = 49), participants with either high or low AI-literacy evaluated all four methods in terms of explanation satisfaction, cognitive load, and overall evaluation in the same classification task environment. High AI-literacy participants were largely indifferent to the four methods, whereas low AI-literacy participants favoured the visual map, perceiving it as the least cognitively demanding. Our findings contribute to the evaluation and development of XAI methods for different types of end-users.

Machine learning (ML) provides designers with a wide range of opportunities to innovate products and services. However, the design discipline struggles to integrate ML knowledge in education and prepare designers to ideate with ML. We propose the Mix & Match Machine Learning toolkit, which provides relevant ML knowledge in the form of tangible tokens and a web interface to support designers’ ideation processes. The tokens represent data types and ML capabilities. By using the toolkit, designers can explore, understand, combine, and operationalize the capabilities of ML and understand its limitations, without depending on programming or computer science knowledge. We evaluated the toolkit in two workshops with design students, and we found that it supports both learning and ideation goals. We discuss the design implications and potential impact of a hybrid toolkit for ML on design education and practice.

Social media platforms newsfeeds are generated by AI algorithms, which select and order posts based on user data. However, users are often unaware of what data is collected and employed for this aim, neither can they control it. To open up discussions on what data users are willing to feed the newsfeed algorithm with, we created the Facebook Data Shield, a human-size interactive installation where users can see and control what type of data is collected. By pressing buttons, data categories and/or data variables can be (de)activated. An outer rim with lights gives feedback to users about the level of personalization of the resulting newsfeed. We performed a preliminary study to get insights into what data users are willing to share, their preferred level of control, and the effect of such an installation on users’ awareness. Based on our findings, we discuss implications for design and future work.