Millions of people now use non-clinical Large Language Model (LLM) tools like ChatGPT for mental well-being support. This paper investigates what it means to design such tools responsibly, and how to operationalize that responsibility in their design and evaluation. By interviewing experts and analyzing related regulations, we found that designing an LLM tool responsibly involves: (1) Articulating the specific benefits it guarantees and for whom. Does it guarantee specific, proven relief, like an over-the-counter drug, or offer minimal guarantees, like a nutritional supplement? (2) Specifying the LLM tool’s “active ingredients” for improving well-being and whether it guarantees their effective delivery (like a primary care provider) or not (like a yoga instructor). These specifications outline an LLM tool’s pertinent risks, appropriate evaluation metrics, and the respective responsibilities of LLM developers, tool designers, and users. These analogies—LLM tools as supplements, drugs, yoga instructors, and primary care providers—can scaffold further conversations about their responsible design.

Bystander intervention, or upstanding, is an effective antidote to cyberbullying, but entails many challenges (e.g. self-efficacy, not knowing what to do or how to upstand). Through two studies, this paper investigates collaborative upstanding, examining how a conversational partner (human or AI) can guide bystanders through these challenges in-situ. In a paired role-play study (n=24), we found that bystanders faced significant challenges in how to intervene. Even after deciding to act, how-to challenges often reignited doubts about their self-efficacy and responsibility. Using these insights, we designed ConCUR, a chatbot that (1) encourages bystanders to co-author an upstanding message, leading them to confront how-to challenges sooner, and (2) addresses how-to challenges simultaneously with other challenges that are introduced through a flexible process. Our second study (n=20) suggests such a chatbot is effective in promoting upstanding behavior in the lab setting. We discuss the implications of in-situ collaborative upstanding to upstanding education research, framing upstanding as an iterative and flexible process rather than sequential.

Emerging adults (EAs) often struggle with their identity, making them vulnerable to mental health issues.This paper examines how EAs explore their identity through life-story writing, with an eye on how AI might help.Our interview study found that mandatory writing assignments, such as a Statement of Purpose for college applications, often triggered EAs' identity exploration.These writing/identity exploration processes were collaborative between EAs and those closest to them.Collaborations succeeded when both parties had the skills and confidence to discuss EAs' identities with enough intensity and directness, but not so much that it crossed the boundaries of their relationship.When collaborations failed, EAs resorted to consulting AI.These findings offer an alternative perspective to the traditional design of cognitive AI writing assistants, which assumes writers write about their lives proactively and privately.This work suggests that AI assistants might be more effective if they help (1) initiate EAs' identity reflection while moderating its intensity, and (2) serve as a connective tissue among EAs and their support networks.

In this work, we explore explicit Large Language Model (LLM)powered support for the iterative design of computer programs.Program design, like other design activity, is characterized by navigating a space of alternative problem formulations and associated solutions in an iterative fashion.LLMs are potentially powerful tools in helping this exploration; however, by default, code-generation LLMs deliver code that represents a particular point solution.This obscures the larger space of possible alternatives, many of which might be preferable to the LLM's default interpretation and its generated code.We contribute an IDE that supports program design through generating and showing new ways to frame problems alongside alternative solutions, tracking design decisions, and identifying implicit decisions made by either the programmer or the LLM.In a user study, we find that with our IDE, users combine and parallelize design phases to explore a broader design space-but also struggle to keep up with LLM-originated changes to code and other information overload.These findings suggest a core challenge for future IDEs that support program design through higher-level instructions given to LLM-based agents: carefully managing attention and deciding what information agents should surface to program designers and when.

Natural language processing (NLP) tools have the potential to boost civic participation and enhance democratic processes because they can significantly increase governments' capacity to gather and analyze citizen opinions. However, their adoption in government remains limited, and harnessing their benefits while preventing unintended consequences remains a challenge. While prior work has focused on improving NLP performance, this work examines how different internal government stakeholders influence NLP tools' thoughtful adoption. We interviewed seven politicians (politically appointed officials as heads of government institutions) and thirteen public servants (career government employees who design and administrate policy interventions), inquiring how they choose whether and how to use NLP tools to support civic participation processes. The interviews suggest that policymakers across both groups focused on their needs for career advancement and the need to showcase the legitimacy and fairness of their work when considering NLP tool adoption and use. Because these needs vary between politicians and public servants, their preferred NLP features and tool designs also differ. Interestingly, despite their differing needs and opinions, neither group clearly identifies who should advocate for NLP adoption to enhance civic participation or address the unintended consequences of a poorly considered adoption. This lack of clarity in responsibility might have caused the governments' low adoption of NLP tools. We discuss how these findings reveal new insights for future HCI research. They inform the design of NLP tools for increasing civic participation efficiency and capacity, the design of other tools and methods that ensure thoughtful adoption of AI tools in government, and the design of NLP tools for collaborative use among users with different incentives and needs.

Safety-critical data, such as crash and near-crash records, are crucial to improving autonomous vehicle (AV) design and development. Sharing such data across AV companies, academic researchers, regulators, and the public can help make all AVs safer. However, AV companies rarely share safety-critical data externally. This paper aims to pinpoint why AV companies are reluctant to share safety-critical data, with an eye on how these barriers can inform new approaches to promote sharing. We interviewed twelve AV company employees who actively work with such data in their day-to-day work. Findings suggest two key, previously unknown barriers to data sharing: (1) Datasets inherently embed salient knowledge that is key to improving AV safety and are resource-intensive. Therefore, data sharing, even within a company, is fraught with politics. (2) Interviewees believed AV safety knowledge is private knowledge that brings competitive edges to their companies, rather than public knowledge for social good. We discuss the implications of these findings for incentivizing and enabling safety-critical AV data sharing, specifically, implications for new approaches to (1) debating and stratifying public and private AV safety knowledge, (2) innovating data tools and data sharing pipelines that enable easier sharing of public AV safety data and knowledge ; (3) offsetting costs of curating safety-critical data and incentivizing data sharing.

Cyberbullying harms teenagers’ mental health, and teaching them upstanding intervention is crucial. Wizard-of-Oz studies show chatbots can scale up personalized and interactive cyberbullying education, but implementing such chatbots is a challenging and delicate task. We created a no-code chatbot design tool for K-12 teachers. Using large language models and prompt chaining, our tool allows teachers to prototype bespoke dialogue flows and chatbot utterances. In offering this tool, we explore teachers’ distinctive needs when designing chatbots to assist their teaching, and how chatbot design tools might better support them. Our findings reveal that teachers welcome the tool enthusiastically. Moreover, they see themselves as playwrights guiding both the students’ and the chatbot’s behaviors, while allowing for some improvisation. Their goal is to enable students to rehearse both desirable and undesirable reactions to cyberbullying in a safe environment. We discuss the design opportunities LLM-Chains offer for empowering teachers and the research opportunities this work opens up.

From wearable health tracking to sensor-laden cities, AI-enhanced pervasive sensing platforms promise far-reaching benefits yet also introduce societal risks. How might designers of these platforms effectively navigate their complex ecology and sociotechnical dynamics? To explore this question, we interviewed designers building mental health technologies who undertook this challenge. They are hospital chief medical information officers and startup founders together striving to create new sensors/AI platforms and integrate them into the healthcare ecosystem. We found that, while all designers aspired to build comprehensive care platforms, their efforts focused on serving either consumers or physicians, delivering a subset of healthcare interventions, and demonstrating system effectiveness one metric at a time. Consequently, breakdowns in patient journeys are emerging; societal risks loom large. We describe how the data economy, designers’ mindsets, and evaluation challenges led to these unintended design consequences. We discuss implications for designing pervasive sensing and AI platforms for social good.

Policies significantly shape computation’s societal impact, a crucial HCI concern. However, challenges persist when HCI professionals attempt to integrate policy into their work or affect policy outcomes. Prior research considered these challenges at the “border” of HCI and policy. This paper asks: What if HCI considers policy integral to its intellectual concerns, placing system-people-policy interaction not at the border but nearer the center of HCI research, practice, and education? What if HCI fosters a mosaic of methods and knowledge contributions that blend system, human, and policy expertise in various ways, just like HCI has done with blending system and human expertise? We present this re-imagined HCI-policy relationship as a provocation and highlight its usefulness: It spotlights previously overlooked system-people-policy interaction work in HCI. It unveils new opportunities for HCI’s futuring, empirical, and design projects. It allows HCI to coordinate its diverse policy engagements, enhancing its collective impact on policy outcomes.

Prompting Large Language Models (LLMs) is an exciting new approach to designing chatbots. But can it improve LLM’s user experience (UX) reliably enough to power chatbot products? Our attempt to design a robust chatbot by prompting GPT-3/4 alone suggests: not yet. Prompts made achieving “80%” UX goals easy, but not the remaining 20%. Fixing the few remaining interaction breakdowns resembled herding cats: We could not address one UX issue or test one design solution at a time; instead, we had to handle everything everywhere all at once. Moreover, because no prompt could make GPT reliably say “I don’t know” when it should, the user-GPT conversations had no guardrails after a breakdown occurred, often leading to UX downward spirals. These risks incentivized us to design highly prescriptive prompts and scripted bots, counter to the promises of LLM-powered chatbots. This paper describes this case study, unpacks prompting’s fickleness and its impact on UX design processes, and discusses implications for LLM-based design methods and tools.

Clinical decision support tools (DSTs), powered by Artificial Intelligence (AI), promise to improve clinicians’ diagnostic and treatment decision-making. However, no AI model is always correct. DSTs must enable clinicians to validate each AI suggestion, convincing them to take the correct suggestions while rejecting its errors. While prior work often tried to do so by explaining AI’s inner workings or performance, we chose a different approach: We investigated how clinicians validated each other’s suggestions in practice (often by referencing scientific literature) and designed a new DST that embraces these naturalistic interactions. This design uses GPT-3 to draw literature evidence that shows the AI suggestions’ robustness and applicability (or the lack thereof). A prototyping study with clinicians from three disease areas proved this approach promising. Clinicians’ interactions with the prototype also revealed new design and research opportunities around (1) harnessing the complementary strengths of literature-based and predictive decision supports; (2) mitigating risks of de-skilling clinicians; and (3) offering low-data decision support with literature.

Pre-trained large language models (“LLMs”) like GPT-3 can engage in fluent, multi-turn instruction-taking out-of-the-box, making them attractive materials for designing natural language interactions. Using natural language to steer LLM outputs (“prompting”) has emerged as an important design technique potentially accessible to non-AI-experts. Crafting effective prompts can be challenging, however, and prompt-based interactions are brittle. Here, we explore whether non-AI-experts can successfully engage in “end-user prompt engineering” using a design probe—a prototype LLM-based chatbot design tool supporting development and systematic evaluation of prompting strategies. Ultimately, our probe participants explored prompt designs opportunistically, not systematically, and struggled in ways echoing end-user programming systems and interactive machine learning systems. Expectations stemming from human-to-human instructional experiences, and a tendency to overgeneralize, were barriers to effective prompt design. These findings have implications for non-AI-expert-facing LLM-based tool design and for improving LLM-and-prompt literacy among programmers and the public, and present opportunities for further research.

Large language models (LMs) offer unprecedented language generation capabilities and exciting opportunities for interaction design. However, their highly context-dependent capabilities are difficult to grasp and are often subjectively interpreted. In this paper, we argue that by curating and analyzing large interaction datasets, the HCI community can foster more incisive examinations of LMs’ generative capabilities. Exemplifying this approach, we present CoAuthor, a dataset designed for revealing GPT-3’s capabilities in assisting creative and argumentative writing. CoAuthor captures rich interactions between 63 writers and four instances of GPT-3 across 1445 writing sessions. We demonstrate that CoAuthor can address questions about GPT-3’s language, ideation, and collaboration capabilities, and reveal its contribution as a writing “collaborator” under various definitions of good collaboration. Finally, we discuss how this work may facilitate a more principled discussion around LMs’ promises and pitfalls in relation to interaction design. The dataset and an interface for replaying the writing sessions are publicly available at https://coauthor.stanford.edu.

Geographically distributed teams often face challenges in coordination and collaboration, lowering their productivity. Understanding the relationship between team dispersion and productivity is critical for supporting such teams. Extensive prior research has studied these relations in lab settings or using qualitative measures. This paper extends prior work by contributing an empirical case study in a real-world organization, using quantitative measures. We studied 117 new research project teams from the same discipline within an industrial research lab for 6 months. During this time, all teams shared one goal: submitting research papers to the same target conference. We analyzed these teams' dispersion-related characteristics as well as team productivity. Interestingly, we found little statistical evidence that geographic and time differences relate to team productivity. However, organizational and functional distances are predictive of the productivity of the dispersed teams we studied. We discuss the open research questions these findings revealed and their implications for future research.

Artificial Intelligence (AI) plays an increasingly important role in improving HCI and user experience. Yet many challenges persist in designing and innovating valuable human-AI interactions. For example, AI systems can make unpredictable errors, and these errors damage UX and even lead to undesired societal impact. However, HCI routinely grapples with complex technologies and mitigates their unintended consequences. What makes AI different? What makes human-AI interaction appear particularly difficult to design? This paper investigates these questions. We synthesize prior research, our own design and research experience, and our observations when teaching human-AI interaction. We identify two sources of AI's distinctive design challenges: 1) uncertainty surrounding AI's capabilities, 2) AI's output complexity, spanning from simple to adaptive complex. We identify four levels of AI systems. On each level, designers encounter a different subset of the design challenges. We demonstrate how these findings reveal new insights for designers, researchers, and design tool makers in productively addressing the challenges of human-AI interaction going forward.

Personalization of user experience has a long history of success in the HCI community. More recently the community has focused on adaptive user interfaces, supported by machine learning, that reduce interaction efforts and improves user experience by collapsing transactions and pre-filtering results. However, generally, these more recent results have only been demonstrated in the laboratory environment. In this paper, we share the case of a deployed mobile transit app that adapts based on users’ previous usage. We examine the impact of adaptation, both good and bad, and user abandonment rates. We conducted an 18-month assessment where 2,616 participants (with and without vision impairments) were recruited and participated in an A/B study. Finally, we draw some insights on some unusual effects that appear over the long term.

From healthcare to criminal justice, artificial intelligence (AI) is increasingly supporting high-consequence human decisions. This has spurred the field of explainable AI (XAI). This paper seeks to strengthen empirical application-specific investigations of XAI by exploring theoretical underpinnings of human decision making, drawing from the fields of philosophy and psychology. In this paper, we propose a conceptual framework for building human-centered, decision-theory-driven XAI based on an extensive review across these fields. Drawing on this framework, we identify pathways along which human cognitive patterns drives needs for building XAI and how XAI can mitigate common cognitive biases. We then put this framework into practice by designing and implementing an explainable clinical diagnostic tool for intensive care phenotyping and conducting a co-design exercise with clinicians. Thereafter, we draw insights into how this framework bridges algorithm-generated explanations and human decision-making theories. Finally, we discuss implications for XAI design and development.

This paper investigates how to sketch NLP-powered user experiences. Sketching is a cornerstone of design innovation. When sketching, designers rapidly experiment with a number of abstract ideas using simple, tangible instruments such as drawings and paper prototypes. Sketching NLP-powered experiences, however, presents challenges, i.e. How to visualize abstract language interaction? How to ideate a broad range of technically feasible intelligent functionalities? As a first step towards understanding these challenges, we present a first-person account of our sketching process when designing intelligent writing assistance. We detail the challenges we encountered and emergent solutions, such as a new format of wireframe for sketching language interactions and a new wizard-of-oz-based NLP rapid prototyping method. Drawing on these findings, we discuss the importance of abstraction in sketching and other implications.

Clinical decision support tools (DST) promise improved healthcare outcomes by offering data-driven insights. While effective in lab settings, almost all DSTs have failed in practice. Empirical research diagnosed poor contextual fit as the cause. This paper describes the design and field evaluation of a radically new form of DST. It automatically generates slides for clinicians' decision meetings with subtly embedded machine prognostics. This design took inspiration from the notion of Unremarkable Computing, that by augmenting the users' routines technology/AI can have significant importance for the users yet remain unobtrusive. Our field evaluation suggests clinicians are more likely to encounter and embrace such a DST. Drawing on their responses, we discuss the importance and intricacies of finding the right level of unremarkableness in DST design, and share lessons learned in prototyping critical AI systems as a situated experience.

Machine learning (ML) promises data-driven insights and solutions for people from all walks of life, but the skill of crafting these solutions is possessed by only a few. Emerging research addresses this issue by creating ML tools that are easy and accessible to people who are not formally trained in ML (non-experts). This work investigated how non-experts build ML solutions for themselves in real life. Our interviews and surveys revealed unique potentials of non-expert ML, as well several pitfalls that non-experts are susceptible to. For example, many perceived percentage accuracy as a sole measure of performance, thus problematic models proceeded to deployment. These observations suggested that, while challenging, making ML easy and robust should both be important goals of designing novice-facing ML tools. To advance on this insight, we discuss design implications and created a sensitizing concept to demonstrate how designers might guide non-experts to easily build robust solutions.

Machine learning (ML) plays an increasingly important role in improving a user's experience. However, most UX practitioners face challenges in understanding ML's capabilities or envisioning what it might be. We interviewed 13 designers who had many years of experience designing the UX of ML-enhanced products and services. We probed them to characterize their practices. They shared they do not view themselves as ML experts, nor do they think learning more about ML would make them better designers. Instead, our participants appeared to be the most successful when they engaged in ongoing collaboration with data scientists to help envision what to make and when they embraced a data-centric culture. We discuss the implications of these findings in terms of UX education and as opportunities for additional design research in support of UX designers working with ML.

HCI has become particularly interested in using machine learning (ML) to improve user experience (UX). However, some design researchers claim that there is a lack of design innovation in envisioning how ML might improve UX. We investigate this claim by analyzing 2,494 related HCI research publications. Our review confirmed a lack of research integrating UX and ML. To help span this gap, we mined our corpus to generate a topic landscape, mapping out 7 clusters of ML technical capabilities within HCI. Among them, we identified 3 under-explored clusters that design researchers can dig in and create sensitizing concepts for. To help operationalize these technical design materials, our analysis then identified value channels through which the technical capabilities can provide value for users: self, context, optimal, and utility-capability. The clusters and the value channels collectively mark starting places for envisioning new ways for ML technology to improve people's lives.

With progress in enabling autonomous cars to drive safely on the road, it is time to start asking how they should be driving. A common answer is that they should be adopting their users' driving style. This makes the assumption that users want their autonomous cars to drive like they drive - aggressive drivers want aggressive cars, defensive drivers want defensive cars. In this paper, we put that assumption to the test. We find that users tend to prefer a significantly more defensive driving style than their own. Interestingly, they prefer the style they think is their own, even though their actual driving style tends to be more aggressive. We also find that preferences do depend on the specific driving scenario, opening the door for new ways of learning driving style preference.

Machine learning improves mobile user experience. Interestingly, envisioning apps with adaptive interfaces that reduce navigation and selection effort is not standard UX practice. When implementing an adaptive UI for our mobile transit app, we encountered a number of problems. Our original design did not log necessary information nor did it induce users to provide good labels. On reflection, we realized UX designers should identify and refine UI adaptions when sketching wireframes. To advance on this insight, we reviewed the interfaces of popular apps and extracted six design patterns where UI adaptation can improve in-app navigation. Next, we designed an exemplar set of wireframes, illustrating how UX designers might annotate their interaction flows to communicate planned adaptation and note the information (logs and labels) needed to make the desired inferences.

Clinical decision support tools (DSTs) are computational systems that aid healthcare decision-making. While effective in labs, almost all these systems failed when they moved into clinical practice. Healthcare researchers speculated it is most likely due to a lack of user-centered HCI considerations in the design of these systems. This paper describes a field study investigating how clinicians make a heart pump implant decision with a focus on how to best integrate an intelligent DST into their work process. Our findings reveal a lack of perceived need for and trust of machine intelligence, as well as many barriers to computer use at the point of clinical decision-making. These findings suggest an alternative perspective to the traditional use models, in which clinicians engage with DSTs at the point of making a decision. We identify situations across patients' healthcare trajectories when decision supports would help, and we discuss new forms it might take in these situations.