Abstract
In the discourse on AI regulation, 'responsible AI' is the dominant paradigm, with the focus on mitigating the risks related to AI systems. While this focus is important and necessary, it has limited use for a systematic consideration of AI's societal impact. This paper proposes a proto-framework for assessing the societal impact of AI systems by operationalising the concept of freedom. This proto-framework is intended as a step towards a fully operationalised framework to be used in policymaking contexts. By drawing on Kantian philosophy and related contemporary interpretations, freedom is developed as the counterpart to the concept of responsibility. Two dimensions of freedom are developed in further detail: freedom as capability and freedom as opportunity. These two dimensions of freedom are then applied in a proto-framework that systematically considers AI's impact on society using the Sustainable Development Goals. This proto-framework aims to complement current risk-based approaches and thereby offers a first step towards operationalising the concept of freedom in AI regulation.

Abstract
In the face of increasing austerity and threats of AI-enabled labor replacement at the University of Michigan, a group of workers and students have coalesced around the project of "AI resistance" since Fall 2024. Forming a cross-departmental coalition including librarians, faculty, staff, graduate workers, and undergraduate students, we have hosted a public workshop questioning the techno-deterministic inevitability of AI use at the University and are working with other campus organizations to maintain an ongoing organizing space. This workshop submission incorporates our reflections thus far on the strategies we've employed, the challenges to collective resistance, and our role as workers in resisting AI within the University. Our aim for this work is to provide concrete inspiration for technologists, students, and staff looking to resist AI techno-solutionism within their own universities.

Abstract
The efficacy of AI agents in healthcare research is hindered by their reliance on static, predefined strategies. This creates a critical limitation: agents can become better tool-users but cannot learn to become better strategic planners, a crucial skill for complex domains like healthcare. We introduce HealthFlow, a self-evolving AI agent that overcomes this limitation through a novel meta-level evolution mechanism. HealthFlow autonomously refines its own high-level problem-solving policies by distilling procedural successes and failures into a durable, strategic knowledge base. To anchor our research and facilitate reproducible evaluation, we introduce EHRFlowBench, a new benchmark featuring complex, realistic health data analysis tasks derived from peer-reviewed clinical research. Our comprehensive experiments demonstrate that HealthFlow's self-evolving approach significantly outperforms state-of-the-art agent frameworks. This work marks a necessary shift from building better tool-users to designing smarter, self-evolving task-managers, paving the way for more autonomous and effective AI for scientific discovery.

Abstract
Root Cause Analysis (RCA) is a critical tool for investigating adverse events in healthcare and improving patient safety. However, existing RCA training programs are often limited by high resource demands, leading to insufficient training and inconsistent implementation. To address this challenge, we present an AI-powered 3D simulation game that helps healthcare professionals develop RCA skills through interactive, immersive simulations. This approach offers a cost-effective, scalable, and accessible alternative to traditional training. The prototype simulates an RCA investigation following a death in the ICU, where learners interview five virtual avatars representing ICU team members to investigate the incident and complete a written report. The system enables natural, life-like interactions with avatars via large language models (LLMs), emotional text-to-speech, and AI-powered animations. An additional LLM component provides formative and summative feedback to support continual improvement. We conclude by outlining plans to empirically evaluate the system's efficacy.

Abstract
Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, vision-language-model (VLM) agents can parse webpages, evaluate products, and transact. This raises a fundamental question: what do AI agents buy, and why? We develop ACES, a sandbox environment that pairs a platform-agnostic VLM agent with a fully programmable mock marketplace to study this question. We first conduct basic rationality checks in the context of simple tasks, and then, by randomizing product positions, prices, ratings, reviews, sponsored tags, and platform endorsements, we obtain causal estimates of how frontier VLMs actually shop. Models show strong but heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal "top" rank. They penalize sponsored tags and reward endorsements. Sensitivities to price, ratings, and reviews are directionally human-like but vary sharply in magnitude across models. Motivated by scenarios where sellers use AI agents to optimize product listings, we show that a seller-side agent that makes minor tweaks to product descriptions, targeting AI buyer preferences, can deliver substantial market-share gains if AI-mediated shopping dominates. We also find that modal product choices can differ across models and, in some cases, demand may concentrate on a few select products, raising competition questions. Together, our results illuminate how AI agents may behave in e-commerce settings and surface concrete seller strategy, platform design, and regulatory questions in an AI-mediated ecosystem.

Abstract
Generative AI is no longer a peripheral tool in higher education. It is rapidly evolving into a general-purpose infrastructure that reshapes how knowledge is generated, mediated, and validated. This paper presents findings from a controlled experiment evaluating a Socratic AI Tutor, a large language model designed to scaffold student research question development through structured dialogue grounded in constructivist theory. Conducted with 65 pre-service teacher students in Germany, the study compares interaction with the Socratic Tutor to engagement with an uninstructed AI chatbot. Students using the Socratic Tutor reported significantly greater support for critical, independent, and reflective thinking, suggesting that dialogic AI can stimulate metacognitive engagement and challenging recent narratives of de-skilling due to generative AI usage. These findings serve as a proof of concept for a broader pedagogical shift: the use of multi-agent systems (MAS) composed of specialised AI agents. To conceptualise this, we introduce the notion of orchestrated MAS, modular, pedagogically aligned agent constellations, curated by educators, that support diverse learning trajectories through differentiated roles and coordinated interaction. To anchor this shift, we propose an adapted offer-and-use model, in which students appropriate instructional offers from these agents. Beyond technical feasibility, we examine system-level implications for higher education institutions and students, including funding necessities, changes to faculty roles, curriculars, competencies and assessment practices. We conclude with a comparative cost-effectiveness analysis highlighting the scalability of such systems. In sum, this study contributes both empirical evidence and a conceptual roadmap for hybrid learning ecosystems that embed human-AI co-agency and pedagogical alignment.

Abstract
Large language models (LLMs) can now generate physics practice problems in real time, yet the educational value of these items hinges on rapid, reliable post-generation vetting. We investigated which automated checks are both technically feasible and pedagogically meaningful when exercises are produced on demand within a chatbot interface. A cohort of 34 introductory-physics students generated and attempted 543 problems during exam preparation. Each item was labeled by an expert on a wide range of quality attributes and presented to the learners in pairs to record their preference. We then (i) benchmarked three commodity LLMs as "judges" against the expert labels, (ii) quantified which attributes predict student choice via random-forest models, and (iii) triangulated these results with free-form exit surveys. Only a small subset of the original rubric proved necessary to reliably address student preferences either directly or by proxy. The study demonstrates that scalable formative assessment does not require exhaustive scoring: a carefully curated core of structural and learner-visible checks is sufficient to ensure both technical soundness and user appeal. The findings provide a practical blueprint for deploying real-time, AI-generated practice in physics and other quantitative disciplines.

Abstract
Social determinants are variables that, while not directly pertaining to any specific individual, capture key aspects of contexts and environments that have direct causal influences on certain attributes of an individual. Previous algorithmic fairness literature has primarily focused on sensitive attributes, often overlooking the role of social determinants. Our paper addresses this gap by introducing formal and quantitative rigor into a space that has been shaped largely by qualitative proposals regarding the use of social determinants. To demonstrate theoretical perspectives and practical applicability, we examine a concrete setting of college admissions, using region as a proxy for social determinants. Our approach leverages a region-based analysis with Gamma distribution parameterization to model how social determinants impact individual outcomes. Despite its simplicity, our method quantitatively recovers findings that resonate with nuanced insights in previous qualitative debates, that are often missed by existing algorithmic fairness approaches. Our findings suggest that mitigation strategies centering solely around sensitive attributes may introduce new structural injustice when addressing existing discrimination. Considering both sensitive attributes and social determinants facilitates a more comprehensive explication of benefits and burdens experienced by individuals from diverse demographic backgrounds as well as contextual environments, which is essential for understanding and achieving fairness effectively and transparently.

Abstract
Neurosymbolic AI combines neural network adaptability with symbolic reasoning, promising an approach to address the complex regulatory, operational, and safety challenges in Advanced Air Mobility (AAM). This survey reviews its applications across key AAM domains such as demand forecasting, aircraft design, and real-time air traffic management. Our analysis reveals a fragmented research landscape where methodologies, including Neurosymbolic Reinforcement Learning, have shown potential for dynamic optimization but still face hurdles in scalability, robustness, and compliance with aviation standards. We classify current advancements, present relevant case studies, and outline future research directions aimed at integrating these approaches into reliable, transparent AAM systems. By linking advanced AI techniques with AAM's operational demands, this work provides a concise roadmap for researchers and practitioners developing next-generation air mobility solutions.

Abstract
Air pollution is the world's largest environmental risk factor for human disease and premature death, resulting in more than 6 million permature deaths in 2019. Currently, there is still a challenge to model one of the most important air pollutants, surface ozone, particularly at scales relevant for human health impacts, with the drivers of global ozone trends at these scales largely unknown, limiting the practical use of physics-based models. We employ a 2D Convolutional Neural Network based architecture that estimate surface ozone MOMO-Chem model residuals, referred to as model bias. We demonstrate the potential of this technique in North America and Europe, highlighting its ability better to capture physical model residuals compared to a traditional machine learning method. We assess the impact of incorporating land use information from high-resolution satellite imagery to improve model estimates. Importantly, we discuss how our results can improve our scientific understanding of the factors impacting ozone bias at urban scales that can be used to improve environmental policy.

Abstract
More than a billion people around the world experience intermittence in their water supply, posing challenges for urban households in Global South cities. An intermittent water supply (IWS) system prompts water users to adapt to service deficits which entails coping costs. Adaptation and its impacts can vary between households within the same city, leading to intra-urban inequality. Studies on household adaptation to IWS through survey data are limited to exploring income-based heterogeneity and do not account for the multidimensional and non-linear nature of the data. There is a need for a standardized methodology for understanding household responses to IWS that acknowledges the heterogeneity of households characterized by sets of multiple underlying factors and that is applicable across different settings. Here, we develop an analysis pipeline that applies hierarchical clustering analysis (HCA) in combination with the Welch-two-sample t-test on household survey data from Amman, Jordan. We identify three clusters of households distinguished by a set of characteristics including income, water social network, supply duration, relocation and water quality problems and identify their group-specific adaptive strategies such as contacting the utility or accessing an alternate water source. This study uncovers the unequal nature of IWS adaptation in Amman, giving insights into the link between household characteristics and adaptive behaviors, while proposing a standardized method to reveal relevant heterogeneity in households adapting to IWS.

Abstract
We investigate whether artificial intelligence can autonomously recover known structures of the Standard Model of particle physics using only experimental data and without theoretical inputs. By applying unsupervised machine learning techniques -- including data dimensionality reduction and clustering algorithms -- to intrinsic particle properties and decay modes, we uncover key organizational features of particle physics, such as the relative strength of different interactions and the difference between baryons and mesons. We also identify conserved quantities such as baryon number, strangeness and charm as well as the structure of isospin and the Eightfold Way multiplets. Our analysis then reveals that clustering can separate particles by interaction, flavor symmetries as well as quantum numbers. Additionally, we observe patterns consistent with Regge trajectories in baryon excitations. Our results demonstrate that machine learning can reproduce key aspects of the Standard Model directly from data, suggesting a promising path toward data-driven discovery in fundamental physics.

Abstract
Social intelligence has become a critical capability for large language models (LLMs), enabling them to engage effectively in real-world social tasks such as accommodation, persuasion, collaboration, and negotiation. Reinforcement learning (RL) is a natural fit for training socially intelligent agents because it allows models to learn sophisticated strategies directly through social interactions. However, social interactions have two key characteristics that set barriers for RL training: (1) partial observability, where utterances have indirect and delayed effects that complicate credit assignment, and (2) multi-dimensionality, where behaviors such as rapport-building or knowledge-seeking contribute indirectly to goal achievement. These characteristics make Markov decision process (MDP)-based RL with single-dimensional episode-level rewards inefficient and unstable. To address these challenges, we propose Sotopia-RL, a novel framework that refines coarse episode-level feedback into utterance-level, multi-dimensional rewards. Utterance-level credit assignment mitigates partial observability by attributing outcomes to individual utterances, while multi-dimensional rewards capture the full richness of social interactions and reduce reward hacking. Experiments in Sotopia, an open-ended social learning environment, demonstrate that Sotopia-RL achieves state-of-the-art social goal completion scores (7.17 on Sotopia-hard and 8.31 on Sotopia-full), significantly outperforming existing approaches. Ablation studies confirm the necessity of both utterance-level credit assignment and multi-dimensional reward design for RL training. Our implementation is publicly available at: https://github.com/sotopia-lab/sotopia-rl.

Abstract
In this paper, we cover approaches to systematically govern, assess and quantify bias across the complete life cycle of machine learning models, from initial development and validation to ongoing production monitoring and guardrail implementation. Building upon our foundational work on the Bias Evaluation and Assessment Test Suite (BEATS) for Large Language Models, the authors share prevalent bias and fairness related gaps in Large Language Models (LLMs) and discuss data and AI governance framework to address Bias, Ethics, Fairness, and Factuality within LLMs. The data and AI governance approach discussed in this paper is suitable for practical, real-world applications, enabling rigorous benchmarking of LLMs prior to production deployment, facilitating continuous real-time evaluation, and proactively governing LLM generated responses. By implementing the data and AI governance across the life cycle of AI development, organizations can significantly enhance the safety and responsibility of their GenAI systems, effectively mitigating risks of discrimination and protecting against potential reputational or brand-related harm. Ultimately, through this article, we aim to contribute to advancement of the creation and deployment of socially responsible and ethically aligned generative artificial intelligence powered applications.

Abstract
The conceptual framework proposed in this paper centers on the development of a deliberative moral reasoning system - one designed to process complex moral situations by generating, filtering, and weighing normative arguments drawn from diverse ethical perspectives. While the framework is rooted in Machine Ethics, it also makes a substantive contribution to Value Alignment by outlining a system architecture that links structured moral reasoning to action under time constraints. Grounded in normative moral pluralism, this system is not constructed to imitate behavior but is built on reason-sensitive deliberation over structured moral content in a transparent and principled manner. Beyond its role as a deliberative system, it also serves as the conceptual foundation for a novel two-level architecture: functioning as a moral reasoning teacher envisioned to train faster models that support real-time responsiveness without reproducing the full structure of deliberative reasoning. Together, the deliberative and intuitive components are designed to enable both deep reflection and responsive action. A key design feature is the dual-hybrid structure: a universal layer that defines a moral threshold through top-down and bottom-up learning, and a local layer that learns to weigh competing considerations in context while integrating culturally specific normative content, so long as it remains within the universal threshold. By extending the notion of moral complexity to include not only conflicting beliefs but also multifactorial dilemmas, multiple stakeholders, and the integration of non-moral considerations, the framework aims to support morally grounded decision-making in realistic, high-stakes contexts.

Abstract
Technological advancements have revolutionized numerous industries, including transportation. While digitalization, automation, and connectivity have enhanced safety and efficiency, they have also introduced new vulnerabilities. With 95% of data breaches attributed to human error, promoting cybersecurity awareness in transportation is increasingly critical. Despite numerous cyberattacks on transportation systems worldwide, comprehensive and centralized records of these incidents remain scarce. To address this gap and enhance cyber awareness, this paper presents a large language model (LLM) based approach to extract and organize transportation related cyber incidents from publicly available datasets. A key contribution of this work is the use of generative AI to transform unstructured, heterogeneous cyber incident data into structured formats. Incidents were sourced from the Center for Strategic & International Studies (CSIS) List of Significant Cyber Incidents, the University of Maryland Cyber Events Database (UMCED), the European Repository of Cyber Incidents (EuRepoC), the Maritime Cyber Attack Database (MCAD), and the U.S. DOT Transportation Cybersecurity and Resiliency (TraCR) Examples of Cyber Attacks in Transportation (2018 to 2022). These were classified by a fine tuned LLM into five transportation modes: aviation, maritime, rail, road, and multimodal, forming a transportation specific cyber incident database. Another key contribution of this work is the development of a Retrieval Augmented Generation question answering system, designed to enhance accessibility and practical use by enabling users to query the curated database for specific details on transportation related cyber incidents. By leveraging LLMs for both data extraction and user interaction, this study contributes a novel, accessible tool for improving cybersecurity awareness in the transportation sector.

Abstract
This chapter systematically promotes an emerging interdisciplinary field of human-artificial intelligence interaction (human-AI interaction, HAII) from a human-centered AI (HCAI) perspective. It introduces a framework of human-centered HAII (HC-HAII). HC-HAII places humans at the core of HAII research and applications, emphasizing the importance of adopting a human-centered approach over a technology-centered one. The chapter presents the HC-HAII methodology, including human-centered methods, process, interdisciplinary teams, and multi-level design paradigms. It also highlights key research challenges and future directions. As the first chapter, this chapter also provides a structural overview of this book, which brings together contributions from an interdisciplinary community of researchers and practitioners to advance the theory, methodology, and applications of HCAI in diverse domains of HAII. The purpose of this chapter is to provide a fundamental framework for this book, centered on HAII research and applications based on the HCAI approach, which will pave the way for the content of subsequent chapters.

Abstract
The material footprint of information and communications technology (ICT) systems is both significant and growing, inspiring a variety of conversations around sustainability and climate justice. In part this effort has been catalysed by past scholarship and analysis from the LIMITS community. This paper examines energy storage systems for computing, particularly batteries -- which are discarded at the rate of 15 billion a year worldwide. The International Energy Agency (IEA) is now referring to the energy transition toward low carbon systems as a critical mineral problem, and countries are speaking openly of 'mineral security' in policy documents. In this paper I 1) present a definition for energy experience and what this means for the design and making of devices, interactions and experiences. I also 2) explore a series of electronics device prototypes converted to run from batteryless sustainable energy that are extremely long lasting, and make limited use of critical minerals. As transitional energy experience device-design experiments, what do prototypes like these suggest for more mainstream, mass-manufactured systems?

Abstract
Reducing domestic energy demand is central to climate mitigation and fuel poverty strategies, yet the impact of energy efficiency interventions is highly heterogeneous. Using a causal machine learning model trained on nationally representative data of the English housing stock, we estimate average and conditional treatment effects of wall insulation on gas consumption, focusing on distributional effects across energy burden subgroups. While interventions reduce gas demand on average (by as much as 19 percent), low energy burden groups achieve substantial savings, whereas those experiencing high energy burdens see little to no reduction. This pattern reflects a behaviourally-driven mechanism: households constrained by high costs-to-income ratios (e.g. more than 0.1) reallocate savings toward improved thermal comfort rather than lowering consumption. Far from wasteful, such responses represent rational adjustments in contexts of prior deprivation, with potential co-benefits for health and well-being. These findings call for a broader evaluation framework that accounts for both climate impacts and the equity implications of domestic energy policy.

Abstract
Large Language Models (LLMs) are widely used for code generation. However, commercial models like ChatGPT require significant computing power, which leads to high energy use and carbon emissions. This has raised concerns about their environmental impact. In this study, we evaluate open-source Small Language Models (SLMs) trained explicitly for code generation and compare their performance and energy efficiency against large LLMs and efficient human-written Python code. The goal is to investigate whether SLMs can match the performance of LLMs on certain types of programming problems while producing more energy-efficient code. We evaluate 150 coding problems from LeetCode, evenly distributed across three difficulty levels: easy, medium, and hard. Our comparison includes three small open-source models, StableCode-3B, StarCoderBase-3B, and Qwen2.5-Coder-3B-Instruct, and two large commercial models, GPT-4.0 and DeepSeek-Reasoner. The generated code is evaluated using four key metrics: run-time, memory usage, energy consumption, and correctness. We use human-written solutions as a baseline to assess the quality and efficiency of the model-generated code. Results indicate that LLMs achieve the highest correctness across all difficulty levels, but SLMs are often more energy-efficient when their outputs are correct. In over 52% of the evaluated problems, SLMs consumed the same or less energy than LLMs.