Despite decades of advancements in Artificial Intelligence (AI), fostering appropriate trust in AI systems remains a challenge. Cognitive biases - systematic deviations from rational judgement - profoundly influence human decision-making, and reliance on such “mental shortcuts” can make AI systems appear more or less trustworthy than they really are, often undermining collaboration outcomes. As AI evolves with more sophisticated and persuasive natural language outputs, particularly through Generative AI (GenAI) and Large Language Models (LLMs), these biases may manifest in new and unpredictable ways, calling for their comprehensive examination. This workshop brings together diverse researchers from HCI, human-centred AI, cognitive psychology, interaction design, and related fields to collaboratively explore how cognitive biases influence trust calibration in human-AI interaction and establish a research agenda. We will explore how biases emerge across the human-AI interaction pipeline, what design strategies can mitigate or even harness these heuristics, and what methods are needed to study these dynamics effectively. Through a highly interactive 90-minute session, participants will map out open challenges, brainstorm tensions and solutions, chart future research directions, and share perspectives from their own diverse disciplinary lenses. Through this workshop, we aim to build a shared understanding of how cognitive biases influence trust in evolving AI systems, and derive a forward-looking, bias-aware research agenda that promotes appropriate trust in human-AI interaction.

AI alignment relies on annotator judgments, yet annotation pipelines often treat annotators as interchangeable, obscuring how their social position shapes annotation. We introduce reflexive annotating as a probe that invites crowd workers to reflect on how their positionality informs subjective annotation judgments in a language model alignment context. Through a qualitative study with crowd workers (N = 30), including follow-up interviews (N = 5), we examine how our probe shapes annotators' behaviour, experience, and the situated metadata it elicits. We find that reflexive annotating captures epistemic metadata beyond static demographics by eliciting intersectional reasoning, surfacing positional humility, and nudging viewpoint change. Crucially, we also denote tensions between reflexive engagement and affective demands such as emotional exposure. We discuss the implications of our work for richer value elicitation and alignment practices that treat annotator judgments as situated and selectively integrate positional metadata.

The ideal conversational recommender system (CRS) acts like a savvy salesperson, adapting its language and suggestions to a user's expertise level. However, most current systems treat all users as experts, leading to frustrating and inefficient interactions when users are unfamiliar with a domain. Systems that can adapt their conversational strategies to a user's knowledge level stand to offer a much more natural and effective experience. To enable such adaptation, a CRS must first be able to estimate a user's domain knowledge from interaction signals. Yet, accurately estimating knowledge typically requires tailored interactions to elicit those signals in the first place, creating a fundamental chicken-and-egg problem. In this work, we take a first step toward breaking this dependency by introducing a new task: estimating user domain knowledge directly from conversational transcripts. A key obstacle to such estimation is the lack of suitable data; to our knowledge, no existing dataset captures the conversational behaviors of users with varying levels of domain knowledge. Furthermore, in most dialogue collection protocols, users are free to express their own preferences, which tends to concentrate on popular items and well-known features, offering little insight into how novices explore or learn about unfamiliar features. To address this, we design RecQuest, a game-with-a-purpose data collection protocol that elicits varied expressions of knowledge while using a target-aware CRS to guide interactions, release the resulting dataset, and provide baseline methods and analyses to support future work on user-knowledge-aware CRS.

As AI systems are increasingly adopted in high-stakes domains such as healthcare, autonomous driving, and criminal justice, their failures may threaten human safety and rights. Human oversight of AI systems is therefore critically important, as a potential safeguard to prevent harmful consequences in high-risk AI applications. Although regulations like the European AI Act mandate human oversight for high-risk AI, we lack methodologies and conceptual clarity to implement it effectively. Independent of policy and regulation, poorly designed oversight can create dangerous illusions of safety while obscuring accountability. This interdisciplinary workshop aims to bring together researchers from various disciplines, including AI, HCI, psychology, law, and policy, to address this critical gap. We will explore the following questions — How can we design AI systems that enable meaningful human oversight? What methods effectively communicate system states and risks to human overseers? How do we ensure scalable and effective interventions? Through papers, talks, and interactive group discussions, participants will identify oversight challenges, examine stakeholder roles, discuss supporting tools, methods, regulatory frameworks, and establish a collaborative research agenda. Our central goal is to further a roadmap that enables effective human oversight for the responsible deployment of AI in society.

Privacy concerns significantly impact AI adoption, yet little is known about how information environments shape user responses to data leak threats. We conducted a 2 × 3 between-subjects experiment (N = 610) examining how risk versus ambiguity about privacy leaks affects the adoption of AI personalization. Participants chose between standard and AI-personalized product baskets, with personalization requiring data sharing that could leak to pricing algorithms. Under risk (30% leak probability), we found no difference in AI adoption between privacy-threatening and neutral conditions (ca. 50% adoption). Under ambiguity (10-50% range), privacy threats significantly reduced adoption compared to neutral conditions. This effect holds for sensitive demographic data as well as anonymized preference data. Users systematically over-bid for privacy disclosure labels, suggesting strong demand for transparency institutions. Notably, privacy leak threats did not affect subsequent bargaining behavior with algorithms. Our findings indicate that ambiguity over data leaks, rather than only privacy preferences per se, drives avoidance behavior among users towards personalized AI.

AI consumer markets are characterized by severe buyer-supplier market asymmetries. Complex AI systems can appear highly accurate while making costly errors or embedding hidden defects. While there have been regulatory efforts surrounding different forms of disclosure, large information gaps remain. This paper provides the first experimental evidence on the important role of information asymmetries and disclosure designs in shaping user adoption of AI systems. We systematically vary the density of low-quality AI systems and the depth of disclosure requirements in a simulated AI product market to gauge how people react to the risk of accidentally relying on a low-quality AI system. Then, we compare participants' choices to a rational Bayesian model, analyzing the degree to which partial information disclosure can improve AI adoption. Our results underscore the deleterious effects of information asymmetries on AI adoption, but also highlight the potential of partial disclosure designs to improve the overall efficiency of human decision-making.

While mental health apps can help to promote adolescents’ mental health, prevent mental health problems, and reduce symptoms, maintaining sufficient user engagement with these apps remains challenging. This is often caused by a mismatch between the needs and preferences of adolescents and what the apps offer. Therefore, we need a better understanding of (i) adolescents’ needs and preferences and (ii) potential differences based on user characteristics. To this end, we qualitatively and quantitatively analyzed a dataset describing the user experience of 1312 Dutch adolescents (12–25 years) from the general population after they interacted for several weeks with a gamified mHealth app (the Grow It! app) that aims to promote momentary emotional awareness, reflection, and adaptive coping. A total of 4833 free-text survey responses spanning five user experience survey questions were analyzed using an inductive and iterative coding process, while accounting for intercoder reliability. We used (i) a thematic analysis to identify adolescents’ needs and preferences related to the app, and (ii) an exploratory quantitative analysis of the subthemes to investigate potential differences in which needs and preferences were mentioned by adolescents based on demographics. Through our thematic analysis, we identified three overarching themes related to the app’s design: usability , psychological impact , and meaningful interactive features . Furthermore, we identified two overarching themes that related to the adolescents’ motivation to use the app: intrinsic (de)motivators , and social–environmental factors impacting usage . Each of these themes consisted of four subthemes. Our exploratory statistical analysis shed light on several differences in how frequently these subthemes were mentioned based on age, sex, and educational level. By synthesizing our insights, we identify five design implications that can help tailor future mHealth apps to adolescents’ needs and preferences. These include concrete suggestions to personalize self-monitoring, include actionable insights, align content with personal needs, implement meaningful interactive features (e.g., competitions, gamification, and social communication), and make apps appealing to the entire target group.

Illegal content reporting mechanisms are a key technical and organizational measure through which online platforms address the dissemination of illegal content under European Union law. Under the Digital Services Act (DSA), user notices submitted pursuant to Article 16 must be sufficiently substantiated and provided in good faith, requiring users to interpret legal and procedural language and translate it into legally meaningful categories and reasons. In practice, however, reporting illegal content remains cumbersome across major social media platforms, placing substantial cognitive and legal demands on users. Without effective support at the reporting interface, operationalizing Article 16 in practice remains challenging. We investigate how large language model (LLM)-based assistants can support illegal content reporting. In a controlled user study (N = 450) using an interface modeled on a major platform's reporting workflow, we compare three conditions: (1) a conventional explainable AI assistant (XAI) that suggests a single legal category with a rationale, (2) an evaluative AI assistant (EvalAI) that presents balanced pro and con arguments across candidate legal provisions for user deliberation, and (3) a baseline reflecting unaided reporting (Baseline). We further examine these assistance forms under systematically varied AI error regimes. Our results show that EvalAI improves provision-level accuracy under AI error regimes and reduces misclassification distance relative to conventional XAI, particularly for near-miss and overbreadth errors. In contrast, conventional XAI does not improve - and can degrade - the quality of users' rationales relative to unaided reporting, despite enabling faster decisions when the AI output is correct. We discuss implications for the design of compliance-oriented reporting interfaces, highlighting trade-offs between accuracy, deliberation, and vulnerability to misleading AI output.

Large language models (LLMs) increasingly support heterogeneous tasks within a single interface, requiring users to form, update, and act upon beliefs about one system across domains with different reliability profiles. Understanding how such beliefs transfer across tasks and shape delegation is therefore critical for the design of multipurpose AI systems. We report a preregistered experiment (N = 240, 7,200 trials) in which participants interacted with a controlled AI simulation across grammar checking, travel planning, and visual question answering, each with fixed, domain-typical accuracy levels. Delegation was operationalized as a binary reliance decision - accepting the AI's output versus acting independently and belief dynamics were evaluated against Bayesian benchmarks. We find three main results. First, participants do not reset beliefs between tasks: priors in a new task depend on posteriors from the previous task, with a 10-point increase predicting a 3-4 point higher subsequent prior. Second, within tasks, belief updating follows the Bayesian direction but is substantially conservative, proceeding at roughly half the normative Bayesian rate. Third, delegation is driven primarily by subjective beliefs about AI accuracy rather than self-confidence, though confidence independently reduces reliance when beliefs are held constant. Together, these findings show that users form global, path-dependent expectations about multipurpose AI systems, update them conservatively, and rely on AI primarily based on subjective beliefs rather than objective performance. We discuss implications for expectation calibration, reliance design, and the risks of belief spillovers in deployed LLM-based interfaces.

Conversational recommender systems (CRSs) provide users with an interactive means to express preferences and receive real-time personalized recommendations. The success of these systems is heavily influenced by the preference elicitation process. While existing research mainly focuses on what questions to ask during preference elicitation, there is a notable gap in understanding what role broader interaction patterns - including tone, pacing, and level of proactiveness - play in supporting users in completing a given task. This study investigates the impact of different conversational styles on preference elicitation, task performance, and user satisfaction with CRSs. We conducted a controlled experiment in the context of scientific literature recommendation, contrasting two distinct conversational styles - high involvement (fast-paced, direct, and proactive with frequent prompts) and high considerateness (polite and accommodating, prioritizing clarity and user comfort) - alongside a flexible experimental condition where users could switch between the two. Our results indicate that adapting conversational strategies based on user expertise and allowing flexibility between styles can enhance both user satisfaction and the effectiveness of recommendations in CRSs. Overall, our findings hold important implications for the design of future CRSs.

Since the explosion in popularity of ChatGPT, large language models (LLMs) have continued to impact our everyday lives. Equipped with external tools that are designed for a specific purpose (e.g., for flight booking or an alarm clock), LLM agents exercise an increasing capability to assist humans in their daily work. Although LLM agents have shown a promising blueprint as daily assistants, there is a limited understanding of how they can provide daily assistance based on planning and sequential decision making capabilities. We draw inspiration from recent work that has highlighted the value of g'LLM-modulo' setups in conjunction with humans-in-the-loop for planning tasks. We conducted an empirical study (N = 248) of LLM agents as daily assistants in six commonly occurring tasks with different levels of risk typically associated with them (e.g., flight ticket booking and credit card payments). To ensure user agency and control over the LLM agent, we adopted LLM agents in a plan-then-execute manner, wherein the agents conducted step-wise planning and step-by-step execution in a simulation environment. We analyzed how user involvement at each stage affects their trust and collaborative team performance. Our findings demonstrate that LLM agents can be a double-edged sword - (1) they can work well when a high-quality plan and necessary user involvement in execution are available, and (2) users can easily mistrust the LLM agents with plans that seem plausible. We synthesized key insights for using LLM agents as daily assistants to calibrate user trust and achieve better overall task outcomes. Our work has important implications for the future design of daily assistants and human-AI collaboration with LLM agents.

Recent advances in generative AI have precipitated a proliferation of novel writing assistants. These systems typically rely on multilingual large language models (LLMs), providing globalized workers the ability to revise or create diverse forms of content in different languages. However, there is substantial evidence indicating that the performance of multilingual LLMs varies between languages. Users who employ writing assistance for multiple languages are therefore susceptible to disparate output quality. Importantly, recent research has shown that people tend to generalize algorithmic errors across independent tasks, violating the behavioral axiom of choice independence. In this paper, we analyze whether user utilization of novel writing assistants in a charity advertisement writing task is affected by the AI's performance in a second language. Furthermore, we quantify the extent to which these patterns translate into the persuasiveness of generated charity advertisements, as well as the role of peoples' beliefs about LLM utilization for their donation choices. Our results provide evidence that writers who engage with an LLM-based writing assistant violate choice independence, as prior exposure to a Spanish LLM reduces subsequent utilization of an English LLM. While these patterns do not affect the aggregate persuasiveness of the generated advertisements, people's beliefs about the source of an advertisement (human versus AI) do. In particular, Spanish-speaking female participants who believed that they read an AI-generated advertisement strongly adjusted their donation behavior downwards. Furthermore, people are generally not able to adequately differentiate between human-generated and LLM-generated ads. Our work has important implications on the design, development, integration, and adoption of multilingual LLMs as assistive agents - particularly in writing tasks.

Crowdsourcing marketplaces have gradually flourished over the last decade. With the growing landscape of online work in general, and the rise of paid microtask crowdsourcing in particular, the health and wellbeing of crowd workers has become an important concern. In this paper, we present an online conversational survey, named HealthInsights, for understanding the status quo of workers’ health-related background, physical health, mental health, and their needs. We carried out a study on two popular platforms - Mechanical Turk and Prolific. Results show that the survey has acceptable reliability and validity. We found that workers across these platforms reported similar health-related issues, but also exhibited certain differences. Based on our findings, we argue that crowdsourcing platforms, task requesters, and academic researchers need to take the collective responsibility of creating better work environments. Our work has important implications on task and workflow design that are centered around worker health on crowdsourcing platforms.

Contestability has been proposed as a key element in designing algorithmic decision-making processes that safeguard decision subjects' rights to dignity and autonomy. However, little is known about how contestability can be operationalized based on decision subjects' needs and preferences. We address this research gap by identifying decision subjects' information and procedural needs for enacting meaningful contestability. To this end, we chose an illegal holiday rental detection scenario as our case; a high-risk decision-making process in the public sector. We conducted 21 semi-structured interviews with citizens with experience renting their homes out and different levels of AI literacy. We found that decision subjects request interventions that facilitate (1) cooperation in sense-making, (2) support in contestation acts, and (3) appropriate responsibility attribution. Our results highlight the cooperative work behind contestability, and motivate future efforts to structure individual and collective action, to personalize explanations for contestability, and to open up sites of contestation in AI pipelines.