Artificial intelligence (AI) holds tremendous potential but also poses consequential risks. Regulation frameworks like the EU AI Act aim to mitigate these risks, yet organizations struggle to understand and operationalize Responsible AI (RAI). We introduce the RAI Organizational Maturity (RAI-OM) framework as an initial step towards a RAI maturity model to highlight the many factors that influence an organization's RAI maturity. Developed through in-depth qualitative interviews and co-design sessions with 90 RAI experts, the RAI-OM framework consists of 24 dimensions grouped into three main categories: Organizational Foundations, Team Approach, and RAI Practices. Our findings also provide further evidence for the interdependent nature of RAI's organizational factors, the import of collaboration for mature RAI, and the need to start RAI early in the AI lifecyle. Researchers and practitioners can use the RAI-OM framework and our research findings to not only understand the different moving parts in RAI's complex organizational machinery, but also address organizational barriers to RAI, unpack the different types of collaborations needed for mature RAI, and support RAI's articulation work and process.

Motivations : Recent research has emerged on generally how to improve AI products’ human-AI interaction (HAI) user experience (UX), but relatively little is known about HAI-UX inclusivity. For example, what kinds of users are supported, and who are left out? What product changes would make it more inclusive? Objectives : To help fill this gap, we present an approach to measuring what kinds of diverse users an AI product leaves out and how to act upon that knowledge. To bring actionability to the results, the approach focuses on users’ problem-solving diversity. Thus, our specific objectives were (1) to show how the measure can reveal which participants with diverse problem-solving styles were left behind in a set of AI products and (2) to relate participants’ problem-solving diversity to their demographic diversity, specifically gender and age. Methods : We performed 18 experiments, discarding two that failed manipulation checks. Each experiment was a 2 \(\times\) 2 factorial experiment with online participants, comparing two AI products: one deliberately violating 1 of 18 HAI guidelines and the other applying the same guideline. For our first objective, we used our measure to analyze how much each AI product gained/lost HAI-UX inclusivity compared to its counterpart, where inclusivity meant supportiveness to participants with particular problem-solving styles. For our second objective, we analyzed how participants’ problem-solving styles aligned with their gender identities and ages. Results and Implications : Participants’ diverse problem-solving styles revealed six types of inclusivity results: (1) the AI products that followed an HAI guideline were almost always more inclusive across diversity of problem-solving styles than the products that did not follow that guideline—but “who” got most of the inclusivity varied widely by guideline and by problem-solving style; (2) when an AI product had risk implications, four variables’ values varied in tandem: participants’ feelings of control, their (lack of) suspicion, their trust in the product, and their certainty while using the product; (3) the more control an AI product offered users, the more inclusive it was; (4) whether an AI product was learning from “my” data or other people’s affected how inclusive that product was; (5) participants’ problem-solving styles skewed differently by gender and age group; and (6) almost all of the results suggested actions that HAI practitioners could take to improve their products’ inclusivity further. Together, these results suggest that a key to improving the demographic inclusivity of an AI product (e.g., across a wide range of genders, ages) can often be obtained by improving the product’s support of diverse problem-solving styles.

This work contributes a research protocol for evaluating human-AI interaction in the context of specific AI products. The research protocol enables UX and HCI researchers to assess different human-AI interaction solutions and validate design decisions before investing in engineering. We present a detailed account of the research protocol and demonstrate its use by employing it to study an existing set of human-AI interaction guidelines. We used factorial surveys with a 2 × 2 mixed design to compare user perceptions when a guideline is applied versus violated, under conditions of optimal versus sub-optimal AI performance. The results provided both qualitative and quantitative insights into the UX impact of each guideline. These insights can support creators of user-facing AI systems in their nuanced prioritization and application of the guidelines.

Machine learning (ML) interpretability techniques can reveal undesirable patterns in data that models exploit to make predictions-potentially causing harms once deployed. However, how to take action to address these patterns is not always clear. In a collaboration between ML and human-computer interaction researchers, physicians, and data scientists, we develop GAM Changer, the first interactive system to help domain experts and data scientists easily and responsibly edit Generalized Additive Models (GAMs) and fix problematic patterns. With novel interaction techniques, our tool puts interpretability into action-empowering users to analyze, validate, and align model behaviors with their knowledge and values. Physicians have started to use our tool to investigate and fix pneumonia and sepsis risk prediction models, and an evaluation with 7 data scientists working in diverse domains highlights that our tool is easy to use, meets their model editing needs, and fits into their current workflows. Built with modern web technologies, our tool runs locally in users' web browsers or computational notebooks, lowering the barrier to use. GAM Changer is available at the following public demo link: https://interpret.ml/gam-changer.

Data is central to the development and evaluation of machine learning (ML) models. However, the use of problematic or inappropriate datasets can result in harms when the resulting models are deployed. To encourage responsible AI practice through more deliberate reflection on datasets and transparency around the processes by which they are created, researchers and practitioners have begun to advocate for increased data documentation and have proposed several data documentation frameworks. However, there is little research on whether these data documentation frameworks meet the needs of ML practitioners, who both create and consume datasets. To address this gap, we set out to understand ML practitioners' data documentation perceptions, needs, challenges, and desiderata, with the ultimate goal of deriving design requirements that can inform future data documentation frameworks. We conducted a series of semi-structured interviews with 14 ML practitioners at a single large, international technology company. We had them answer a list of questions taken from datasheets for datasets~\citegebru2018datasheets. Our findings show that current approaches to data documentation are largely ad hoc and myopic in nature. Participants expressed needs for data documentation frameworks to be adaptable to their contexts, integrated into their existing tools and workflows, and automated wherever possible. Despite the fact that data documentation frameworks are often motivated from the perspective of responsible AI, participants did not make the connection between the questions that they were asked to answer and their responsible AI implications. In addition, participants often had difficulties prioritizing the needs of dataset consumers and providing information that someone unfamiliar with their datasets might need to know. Based on these findings, we derive seven design requirements for future data documentation frameworks such as more actionable guidance on how the characteristics of datasets might result in harms and how these harms might be mitigated, more explicit prompts for reflection, automated adaptation to different contexts, and integration into ML practitioners' existing tools and workflows.

Running a household requires a large amount of labor, from ensuring multiple bills are paid to organizing important documents. Failure to manage such information can have critical consequences for the financial and psychological well-being of the family; however, little is known about how families manage the full scale of information encountered in the home. In this paper, we introduce family information management (FIM) as a set of overarching practices involved in managing and coordinating household-related information. To understand how families engage in FIM, we conducted in-depth interviews with members of ten families, which included guided tours of their information archives. We found that families struggle to organize, store, retrieve, and share information, and that there are significant socioemotional costs to this work. We propose opportunities for designing technologies to support FIM and argue that, given the numerous challenges and unmet needs, the understudied area of FIM deserves further investment of research and design efforts.

Advances in artificial intelligence (AI) frame opportunities and challenges for user interface design. Principles for human-AI interaction have been discussed in the human-computer interaction community for over two decades, but more study and innovation are needed in light of advances in AI and the growing uses of AI technologies in human-facing applications. We propose 18 generally applicable design guidelines for human-AI interaction. These guidelines are validated through multiple rounds of evaluation including a user study with 49 design practitioners who tested the guidelines against 20 popular AI-infused products. The results verify the relevance of the guidelines over a spectrum of interaction scenarios and reveal gaps in our knowledge, highlighting opportunities for further research. Based on the evaluations, we believe the set of design guidelines can serve as a resource to practitioners working on the design of applications and features that harness AI technologies, and to researchers interested in the further development of human-AI interaction design principles.

In recent years, research has revealed gender biases in numerous software products. But although some researchers have found ways to improve gender participation in specific software projects, general methods focus mainly on detecting gender biases -- not fixing them. To help fill this gap, we investigated whether the GenderMag bias detection method can lead directly to designs with fewer gender biases. In our 3-step investigation, two HCI researchers analyzed an industrial software product using GenderMag; we derived design changes to the product using the biases they found; and ran an empirical study of participants using the original product versus the new version. The results showed that using the method in this way did improve the software's inclusiveness: women succeeded more often in the new version than in the original; men's success rates improved too; and the gender gap entirely disappeared.

Mobile application designers use onboarding task flows to help first time users learn and engage with key application functionality. Although some guidelines for designing onboarding flows have been offered by practitioners, a systematic, research-informed approach is needed. In this paper, we present the creation of a method for designing mobile application onboarding experiences. We used the minimalist instruction framework to engage twelve university students in an iterative set of design and evaluation activities. Participants interacted with a physical prototype of an educational badging mobile application through a semi-structured exploration and reflection activity, bookended by structured mini-interviews. We found that this method facilitated engagement with participants' meaning-making processes, resulting in useful design insights and the creation of an onboarding task flow. Research opportunities for integrating instructional design and learning approaches in HCI in the context of onboarding are considered.

The rapid growth of the UX profession has led to an increased need for qualified practitioners and a proliferation of UX educational programs offered in both academia and industry. In this note, we present the design and initial evaluation of a new studio-based undergraduate program in UX--the first of its kind at a large, research-intensive US university. The program includes several curricular innovations, such as an integrated studio pedagogy in which six topical strands are interwoven across two types of studios. These studios are interconnected and span five semesters of the undergraduate experience. We present the curriculum model and the foundational principles that informed its design. We describe the two types of studios and their interconnection, and present early evaluation data showing that students are building valuable skills. The program described in this note provides a trailblazing model for UX pedagogy at the undergraduate level.

The main goal of the US Science Gateways Community Institute (SGCI) is to serve science gateways to achieve sustainability and growth. Science gateways allow science and engineering communities to access shared data, software, computing services, instruments, educational materials, and other resources specific to their disciplines. Thus, science gateways are a subgroup of scientific software and the means for addressing software sustainability are also suitable for science gateways and vice versa, e.g., best practices for software engineering. Since science gateways are tailored to specific communities, understanding users' requirements is critical for sustainability. SGCI consists of five service areas that closely interact with each other. The Incubator acknowledges the value of business strategy to inform well-designed science gateways and offers two main types of services: individualized consultancy, tailored to specific challenges a gateway faces, and the Science Gateways Bootcamp. The cornerstone of the Bootcamp is a one-week onsite intensive workshop where participants create their own roadmap for a sustainable science gateway via sessions with experts, hands-on exercises, and group work. This paper offers an overview of the work of the Incubator and shares lessons learned from the inaugural session of the Bootcamp in April 2017.

We present a design study of the Deep Insights Anywhere, Anytime (DIA2) platform, a web-based visual analytics system that allows program managers and academic staff at the U.S. National Science Foundation to search, view, and analyze their research funding portfolio. The goal of this system is to facilitate users' understanding of both past and currently active research awards in order to make more informed decisions of their future funding. This user group is characterized by high domain expertise yet not necessarily high literacy in visualization and visual analytics-they are essentially casual experts-and thus require careful visual and information design, including adhering to user experience standards, providing a self-instructive interface, and progressively refining visualizations to minimize complexity. We discuss the challenges of designing a system for casual experts and highlight how we addressed this issue by modeling the organizational structure and workflows of the NSF within our system. We discuss each stage of the design process, starting with formative interviews, prototypes, and finally live deployments and evaluation with stakeholders.