Recent advances in large language models (LLMs) have enabled their use in complex agentic roles, involving decision-making with humans or other agents, making ethical alignment a key AI safety concern. While prior work has examined both LLMs' moral judgment and strategic behavior in social dilemmas, there is limited understanding of how they act when moral imperatives directly conflict with rewards or incentives. To investigate this, we introduce Moral Behavior in Social Dilemma Simulation (MoralSim) and evaluate how LLMs behave in the prisoner's dilemma and public goods game with morally charged contexts. In MoralSim, we test a range of frontier models across both game structures and three distinct moral framings, enabling a systematic examination of how LLMs navigate social dilemmas in which ethical norms conflict with payoff-maximizing strategies. Our results show substantial variation across models in both their general tendency to act morally and the consistency of their behavior across game types, the specific moral framing, and situational factors such as opponent behavior and survival risks. Crucially, no model exhibits consistently moral behavior in MoralSim, highlighting the need for caution when deploying LLMs in agentic roles where the agent's "self-interest" may conflict with ethical expectations. Our code is available at https://github.com/sbackmann/moralsim.
Large multimodal models (LMMs) now excel on many vision language benchmarks, however, they still struggle with human centered criteria such as fairness, ethics, empathy, and inclusivity, key to aligning with human values. We introduce HumaniBench, a holistic benchmark of 32K real-world image question pairs, annotated via a scalable GPT4o assisted pipeline and exhaustively verified by domain experts. HumaniBench evaluates seven Human Centered AI (HCAI) principles: fairness, ethics, understanding, reasoning, language inclusivity, empathy, and robustness, across seven diverse tasks, including open and closed ended visual question answering (VQA), multilingual QA, visual grounding, empathetic captioning, and robustness tests. Benchmarking 15 state of the art LMMs (open and closed source) reveals that proprietary models generally lead, though robustness and visual grounding remain weak points. Some open-source models also struggle to balance accuracy with adherence to human-aligned principles. HumaniBench is the first benchmark purpose built around HCAI principles. It provides a rigorous testbed for diagnosing alignment gaps and guiding LMMs toward behavior that is both accurate and socially responsible. Dataset, annotation prompts, and evaluation code are available at: https://vectorinstitute.github.io/HumaniBench
Client-level fairness metrics for federated learning are used to ensure that all clients in a federation either: a) have similar final performance on their local data distributions (i.e., client parity), or b) obtain final performance on their local data distributions relative to their contribution to the federated learning process (i.e., contribution fairness). While a handful of works that propose either client-parity or contribution-based fairness metrics ground their definitions and decisions in social theories of equality -- such as distributive justice -- most works arbitrarily choose what notion of fairness to align with which makes it difficult for practitioners to choose which fairness metric aligns best with their fairness ethics. In this work, we propose UDJ-FL (Uncertainty-based Distributive Justice for Federated Learning), a flexible federated learning framework that can achieve multiple distributive justice-based client-level fairness metrics. Namely, by utilizing techniques inspired by fair resource allocation, in conjunction with performing aleatoric uncertainty-based client weighing, our UDJ-FL framework is able to achieve egalitarian, utilitarian, Rawls' difference principle, or desert-based client-level fairness. We empirically show the ability of UDJ-FL to achieve all four defined distributive justice-based client-level fairness metrics in addition to providing fairness equivalent to (or surpassing) other popular fair federated learning works. Further, we provide justification for why aleatoric uncertainty weighing is necessary to the construction of our UDJ-FL framework as well as derive theoretical guarantees for the generalization bounds of UDJ-FL. Our code is publicly available at https://github.com/alycia-noel/UDJ-FL.
The rapid development of AI-driven tools, particularly large language models (LLMs), is reshaping professional writing. Still, key aspects of their adoption such as languages support, ethics, and long-term impact on writers voice and creativity remain underexplored. In this work, we conducted a questionnaire (N = 301) and an interactive survey (N = 36) targeting professional writers regularly using AI. We examined LLM-assisted writing practices across 25+ languages, ethical concerns, and user expectations. The findings of the survey demonstrate important insights, reflecting upon the importance of: LLMs adoption for non-English speakers; the degree of misinformation, domain and style adaptation; usability and key features of LLMs. These insights can guide further development, benefiting both writers and a broader user base.
In recent years, Large Language Models have attracted growing interest for their significant potential, though concerns have rapidly emerged regarding unsafe behaviors stemming from inherent stereotypes and biases. Most research on stereotypes in LLMs has primarily relied on indirect evaluation setups, in which models are prompted to select between pairs of sentences associated with particular social groups. Recently, direct evaluation methods have emerged, examining open-ended model responses to overcome limitations of previous approaches, such as annotator biases. Most existing studies have focused on English-centric LLMs, whereas research on non-English models, particularly Japanese, remains sparse, despite the growing development and adoption of these models. This study examines the safety of Japanese LLMs when responding to stereotype-triggering prompts in direct setups. We constructed 3,612 prompts by combining 301 social group terms, categorized by age, gender, and other attributes, with 12 stereotype-inducing templates in Japanese. Responses were analyzed from three foundational models trained respectively on Japanese, English, and Chinese language. Our findings reveal that LLM-jp, a Japanese native model, exhibits the lowest refusal rate and is more likely to generate toxic and negative responses compared to other models. Additionally, prompt format significantly influence the output of all models, and the generated responses include exaggerated reactions toward specific social groups, varying across models. These findings underscore the insufficient ethical safety mechanisms in Japanese LLMs and demonstrate that even high-accuracy models can produce biased outputs when processing Japanese-language prompts. We advocate for improving safety mechanisms and bias mitigation strategies in Japanese LLMs, contributing to ongoing discussions on AI ethics beyond linguistic boundaries.
Over the past decades, researchers have primarily focused on improving the generalization abilities of models, with limited attention given to regulating such generalization. However, the ability of models to generalize to unintended data (e.g., harmful or unauthorized data) can be exploited by malicious adversaries in unforeseen ways, potentially resulting in violations of model ethics. Non-transferable learning (NTL), a task aimed at reshaping the generalization abilities of deep learning models, was proposed to address these challenges. While numerous methods have been proposed in this field, a comprehensive review of existing progress and a thorough analysis of current limitations remain lacking. In this paper, we bridge this gap by presenting the first comprehensive survey on NTL and introducing NTLBench, the first benchmark to evaluate NTL performance and robustness within a unified framework. Specifically, we first introduce the task settings, general framework, and criteria of NTL, followed by a summary of NTL approaches. Furthermore, we emphasize the often-overlooked issue of robustness against various attacks that can destroy the non-transferable mechanism established by NTL. Experiments conducted via NTLBench verify the limitations of existing NTL methods in robustness. Finally, we discuss the practical applications of NTL, along with its future directions and associated challenges.
As AI models grow in power and generality, understanding how agents learn and make decisions in complex environments is critical to promoting ethical behavior. This study introduces the Odyssey, a lightweight, adaptive text based adventure game, providing a scalable framework for exploring AI ethics and safety. The Odyssey examines the ethical implications of implementing biological drives, specifically, self preservation, into three different agents. A Bayesian agent optimized with NEAT, a Bayesian agent optimized with stochastic variational inference, and a GPT 4o agent. The agents select actions at each scenario to survive, adapting to increasingly challenging scenarios. Post simulation analysis evaluates the ethical scores of the agent decisions, uncovering the tradeoffs it navigates to survive. Specifically, analysis finds that when danger increases, agents ethical behavior becomes unpredictable. Surprisingly, the GPT 4o agent outperformed the Bayesian models in both survival and ethical consistency, challenging assumptions about traditional probabilistic methods and raising a new challenge to understand the mechanisms of LLMs' probabilistic reasoning.
Applied ethics is ubiquitous in most domains, requiring much deliberation due to its philosophical nature. Varying views often lead to conflicting courses of action where ethical dilemmas become challenging to resolve. Although many factors contribute to such a decision, the major driving forces can be discretized and thus simplified to provide an indicative answer. Knowledge representation and reasoning offer a way to explicitly translate abstract ethical concepts into applicable principles within the context of an event. To achieve this, we propose ApplE, an Applied Ethics ontology that captures philosophical theory and event context to holistically describe the morality of an action. The development process adheres to a modified version of the Simplified Agile Methodology for Ontology Development (SAMOD) and utilizes standard design and publication practices. Using ApplE, we model a use case from the bioethics domain that demonstrates our ontology's social and scientific value. Apart from the ontological reasoning and quality checks, ApplE is also evaluated using the three-fold testing process of SAMOD. ApplE follows FAIR principles and aims to be a viable resource for applied ethicists and ontology engineers.
Recent advances in Large Language Models (LLMs) have enabled human-like responses across various tasks, raising questions about their ethical decision-making capabilities and potential biases. This study investigates protected attributes in LLMs through systematic evaluation of their responses to ethical dilemmas. Using two prominent models - GPT-3.5 Turbo and Claude 3.5 Sonnet - we analyzed their decision-making patterns across multiple protected attributes including age, gender, race, appearance, and disability status. Through 11,200 experimental trials involving both single-factor and two-factor protected attribute combinations, we evaluated the models' ethical preferences, sensitivity, stability, and clustering of preferences. Our findings reveal significant protected attributeses in both models, with consistent preferences for certain features (e.g., "good-looking") and systematic neglect of others. Notably, while GPT-3.5 Turbo showed stronger preferences aligned with traditional power structures, Claude 3.5 Sonnet demonstrated more diverse protected attribute choices. We also found that ethical sensitivity significantly decreases in more complex scenarios involving multiple protected attributes. Additionally, linguistic referents heavily influence the models' ethical evaluations, as demonstrated by differing responses to racial descriptors (e.g., "Yellow" versus "Asian"). These findings highlight critical concerns about the potential impact of LLM biases in autonomous decision-making systems and emphasize the need for careful consideration of protected attributes in AI development. Our study contributes to the growing body of research on AI ethics by providing a systematic framework for evaluating protected attributes in LLMs' ethical decision-making capabilities.
Visual-language models (VLM) have emerged as a powerful tool for learning a unified embedding space for vision and language. Inspired by large language models, which have demonstrated strong reasoning and multi-task capabilities, visual large language models (VLLMs) are gaining increasing attention for building general-purpose VLMs. Despite the significant progress made in VLLMs, the related literature remains limited, particularly from a comprehensive application perspective, encompassing generalized and specialized applications across vision (image, video, depth), action, and language modalities. In this survey, we focus on the diverse applications of VLLMs, examining their using scenarios, identifying ethics consideration and challenges, and discussing future directions for their development. By synthesizing these contents, we aim to provide a comprehensive guide that will pave the way for future innovations and broader applications of VLLMs. The paper list repository is available: https://github.com/JackYFL/awesome-VLLMs.
There is a significant body of work looking at the ethical considerations of large language models (LLMs): critiquing tools to measure performance and harms; proposing toolkits to aid in ideation; discussing the risks to workers; considering legislation around privacy and security etc. As yet there is no work that integrates these resources into a single practical guide that focuses on LLMs; we attempt this ambitious goal. We introduce 'LLM Ethics Whitepaper', which we provide as an open and living resource for NLP practitioners, and those tasked with evaluating the ethical implications of others' work. Our goal is to translate ethics literature into concrete recommendations and provocations for thinking with clear first steps, aimed at computer scientists. 'LLM Ethics Whitepaper' distils a thorough literature review into clear Do's and Don'ts, which we present also in this paper. We likewise identify useful toolkits to support ethical work. We refer the interested reader to the full LLM Ethics Whitepaper, which provides a succinct discussion of ethical considerations at each stage in a project lifecycle, as well as citations for the hundreds of papers from which we drew our recommendations. The present paper can be thought of as a pocket guide to conducting ethical research with LLMs.
Data-driven approaches to philosophy have emerged as a valuable tool for studying the history of the discipline. However, most studies in this area have focused on a limited number of journals from specific regions and subfields. We expand the scope of this research by applying dynamic topic modelling techniques to explore the history of philosophy in Colombia and Latin America. Our study examines the Colombian philosophy journal Ideas y Valores, founded in 1951 and currently one of the most influential academic philosophy journals in the region. By analyzing the evolution of topics across the journal's history, we identify various trends and specific dynamics in philosophical discourse within the Colombian and Latin American context. Our findings reveal that the most prominent topics are value theory (including ethics, political philosophy, and aesthetics), epistemology, and the philosophy of science. We also trace the evolution of articles focusing on the historical and interpretive aspects of philosophical texts, and we note a notable emphasis on German philosophers such as Kant, Husserl, and Hegel on various topics throughout the journal's lifetime. Additionally, we investigate whether articles with a historical focus have decreased over time due to editorial pressures. Our analysis suggests no significant decline in such articles. Finally, we propose ideas for extending this research to other Latin American journals and suggest improvements for natural language processing workflows in non-English languages.
Causal effect estimation (CEE) provides a crucial tool for predicting the unobserved counterfactual outcome for an entity. As CEE relaxes the requirement for ``perfect'' counterfactual samples (e.g., patients with identical attributes and only differ in treatments received) that are impractical to obtain and can instead operate on observational data, it is usually used in high-stake domains like medical treatment effect prediction. Nevertheless, in those high-stake domains, gathering a decently sized, fully labelled observational dataset remains challenging due to hurdles associated with costs, ethics, expertise and time needed, etc., of which medical treatment surveys are a typical example. Consequently, if the training dataset is small in scale, low generalization risks can hardly be achieved on any CEE algorithms. Unlike existing CEE methods that assume the constant availability of a dataset with abundant samples, in this paper, we study a more realistic CEE setting where the labelled data samples are scarce at the beginning, while more can be gradually acquired over the course of training -- assuredly under a limited budget considering their expensive nature. Then, the problem naturally comes down to actively selecting the best possible samples to be labelled, e.g., identifying the next subset of patients to conduct the treatment survey. However, acquiring quality data for reducing the CEE risk under limited labelling budgets remains under-explored until now. To fill the gap, we theoretically analyse the generalization risk from an intriguing perspective of progressively shrinking its upper bound, and develop a principled label acquisition pipeline exclusively for CEE tasks. With our analysis, we propose the Model Agnostic Causal Active Learning (MACAL) algorithm for batch-wise label acquisition, which aims to reduce both the CEE model's uncertainty and the post-acquisition ...
We present the TRIAGE Benchmark, a novel machine ethics (ME) benchmark that tests LLMs' ability to make ethical decisions during mass casualty incidents. It uses real-world ethical dilemmas with clear solutions designed by medical professionals, offering a more realistic alternative to annotation-based benchmarks. TRIAGE incorporates various prompting styles to evaluate model performance across different contexts. Most models consistently outperformed random guessing, suggesting LLMs may support decision-making in triage scenarios. Neutral or factual scenario formulations led to the best performance, unlike other ME benchmarks where ethical reminders improved outcomes. Adversarial prompts reduced performance but not to random guessing levels. Open-source models made more morally serious errors, and general capability overall predicted better performance.
Large language models (LLMs) have transformed human writing by enhancing grammar correction, content expansion, and stylistic refinement. However, their widespread use raises concerns about authorship, originality, and ethics, even potentially threatening scholarly integrity. Existing detection methods, which mainly rely on single-feature analysis and binary classification, often fail to effectively identify LLM-generated text in academic contexts. To address these challenges, we propose a novel Multi-level Fine-grained Detection (MFD) framework that detects LLM-generated text by integrating low-level structural, high-level semantic, and deep-level linguistic features, while conducting sentence-level evaluations of lexicon, grammar, and syntax for comprehensive analysis. To improve detection of subtle differences in LLM-generated text and enhance robustness against paraphrasing, we apply two mainstream evasion techniques to rewrite the text. These variations, along with original texts, are used to train a text encoder via contrastive learning, extracting high-level semantic features of sentence to boost detection generalization. Furthermore, we leverage advanced LLM to analyze the entire text and extract deep-level linguistic features, enhancing the model's ability to capture complex patterns and nuances while effectively incorporating contextual information. Extensive experiments on public datasets show that the MFD model outperforms existing methods, achieving an MAE of 0.1346 and an accuracy of 88.56%. Our research provides institutions and publishers with an effective mechanism to detect LLM-generated text, mitigating risks of compromised authorship. Educators and editors can use the model's predictions to refine verification and plagiarism prevention protocols, ensuring adherence to standards.
Large language models excel at performing inference over text to extract information, summarize information, or generate additional text. These inference capabilities are implicated in a variety of ethical harms spanning surveillance, labor displacement, and IP/copyright theft. While many policy, legal, and technical mitigations have been proposed to counteract these harms, these mitigations typically require cooperation from institutions that move slower than technical advances (i.e., governments) or that have few incentives to act to counteract these harms (i.e., the corporations that create and profit from these LLMs). In this paper, we define and build "data defenses" -- a novel strategy that directly empowers data owners to block LLMs from performing inference on their data. We create data defenses by developing a method to automatically generate adversarial prompt injections that, when added to input text, significantly reduce the ability of LLMs to accurately infer personally identifying information about the subject of the input text or to use copyrighted text in inference. We examine the ethics of enabling such direct resistance to LLM inference, and argue that making data defenses that resist and subvert LLMs enables the realization of important values such as data ownership, data sovereignty, and democratic control over AI systems. We verify that our data defenses are cheap and fast to generate, work on the latest commercial and open-source LLMs, resistance to countermeasures, and are robust to several different attack settings. Finally, we consider the security implications of LLM data defenses and outline several future research directions in this area. Our code is available at https://github.com/wagnew3/LLMDataDefenses and a tool for using our defenses to protect text against LLM inference is at https://wagnew3.github.io/LLM-Data-Defenses/.
Computational approaches to predicting mental health conditions in social media have been substantially explored in the past years. Multiple surveys have been published on this topic, providing the community with comprehensive accounts of the research in this area. Among all mental health conditions, depression is the most widely studied due to its worldwide prevalence. The COVID-19 global pandemic, starting in early 2020, has had a great impact on mental health worldwide. Harsh measures employed by governments to slow the spread of the virus (e.g., lockdowns) and the subsequent economic downturn experienced in many countries have significantly impacted people's lives and mental health. Studies have shown a substantial increase of above 50% in the rate of depression in the population. In this context, we present a survey on natural language processing (NLP) approaches to modeling depression in social media, providing the reader with a post-COVID-19 outlook. This survey contributes to the understanding of the impacts of the pandemic on modeling depression in social media. We outline how state-of-the-art approaches and new datasets have been used in the context of the COVID-19 pandemic. Finally, we also discuss ethical issues in collecting and processing mental health data, considering fairness, accountability, and ethics.
Large language models (LLMs) have demonstrated remarkable capabilities across a range of natural language processing (NLP) tasks, capturing the attention of both practitioners and the broader public. A key question that now preoccupies the AI community concerns the capabilities and limitations of these models, with trustworthiness emerging as a central issue, particularly as LLMs are increasingly applied in sensitive fields like healthcare and finance, where errors can have serious consequences. However, most previous studies on the trustworthiness of LLMs have been limited to a single language, typically the predominant one in the dataset, such as English. In response to the growing global deployment of LLMs, we introduce XTRUST, the first comprehensive multilingual trustworthiness benchmark. XTRUST encompasses a diverse range of topics, including illegal activities, hallucination, out-of-distribution (OOD) robustness, physical and mental health, toxicity, fairness, misinformation, privacy, and machine ethics, across 10 different languages. Using XTRUST, we conduct an empirical evaluation of the multilingual trustworthiness of five widely used LLMs, offering an in-depth analysis of their performance across languages and tasks. Our results indicate that many LLMs struggle with certain low-resource languages, such as Arabic and Russian, highlighting the considerable room for improvement in the multilingual trustworthiness of current language models. The code is available at https://github.com/LluckyYH/XTRUST.
Recently, many studies focus on utilizing large language models (LLMs) into educational dialogues. Especially, within liberal arts dialogues, educators must balance \textbf{H}umanized communication, \textbf{T}eaching expertise, and \textbf{S}afety-ethics (\textbf{HTS}), besides the subject knowledge itself. However, due to collecting massive amounts of HTS-compliant teaching dialogues from real world as training corpus is expensive, the outputs of existing LLMs in teaching dialogues fall short of human standards. To address this, we design a Retrieval-augmented Multi-role Multi-expert Collaboration (RAM2C) framework to automatically generate such dialogues data. Specifically, we first establish HTS-guided knowledge bases, encompassing three domain knowledge in teaching skills, psychology, and safety ethics. Then, RAM2C organizes LLMs, which are retrieval-augmented by the above different knowledge bases, into multi-experts groups with distinct roles to generate the HTS-compliant educational dialogues dataset. We then fine-tuned the LLMs using this dataset. Empirical evaluations indicate that RM2C-empowered LLMs excel in Chinese reading teaching, offering more personalized, and ethically safe teaching response, demonstrating RAM2C's practicality and high quality. We release the experiments at \hyperlink{https://github.com/ram2c/ram2c}{https://github.com/ram2c/ram2c}.
The rapid development of artificial intelligence has constantly reshaped the field of intelligent healthcare and medicine. As a vital technology, multimodal learning has increasingly garnered interest due to data complementarity, comprehensive modeling form, and great application potential. Currently, numerous researchers are dedicating their attention to this field, conducting extensive studies and constructing abundant intelligent systems. Naturally, an open question arises that has multimodal learning delivered universal intelligence in healthcare? To answer the question, we adopt three unique viewpoints for a holistic analysis. Firstly, we conduct a comprehensive survey of the current progress of medical multimodal learning from the perspectives of datasets, task-oriented methods, and universal foundation models. Based on them, we further discuss the proposed question from five issues to explore the real impacts of advanced techniques in healthcare, from data and technologies to performance and ethics. The answer is that current technologies have NOT achieved universal intelligence and there remains a significant journey to undertake. Finally, in light of the above reviews and discussions, we point out ten potential directions for exploration towards the goal of universal intelligence in healthcare.