Despite AI's promise for addressing global challenges, empirical understanding of AI adoption in mission-driven organizations (MDOs) remains limited. While research emphasizes individual applications or ethical principles, little is known about how resource-constrained, values-driven organizations navigate AI integration across operations. We conducted thematic analysis of semi-structured interviews with 15 practitioners from environmental, humanitarian, and development organizations across the Global North and South contexts. Our analysis examines how MDOs currently deploy AI, what barriers constrain adoption, and how practitioners envision future integration. MDOs adopt AI selectively, with sophisticated deployment in content creation and data analysis while maintaining human oversight for mission-critical applications. When AI's efficiency benefits conflict with organizational values, decision-making stalls rather than negotiating trade-offs. This study contributes empirical evidence that AI adoption in MDOs should be understood as conditional rather than inevitable, proceeding only where it strengthens organizational sovereignty and mission integrity while preserving human-centered approaches essential to their missions.

Abstract: Addressing the core problem of insufficient trustworthiness in industrial fault diagnosis, stemming from the limitations of existing methods -- both traditional and deep learning - based -- in terms of interpretability, generalization, and uncertainty quantification, this paper proposes a trustworthy industrial fault diagnosis architecture, the Hierarchical Cognitive Arbitration Architecture (HCAA), which integrates probabilistic models with Large Language Models (LLMs). The architecture conducts a preliminary analysis via a diagnostic engine based on a Bayesian network and features an LLM - driven cognitive arbitration module with multimodal input capabilities. This module performs expert - level arbitration on the initial diagnosis by analyzing structured features and diagnostic charts, holding the priority to make the final decision upon detecting conflicts. To ensure the reliability of the system's output, the architecture integrates a confidence calibration module based on Temperature Scaling and a risk assessment module, which objectively quantify system trustworthiness using metrics like Expected Calibration Error (ECE). Experimental results on a dataset containing multiple fault types demonstrate that the proposed framework improves diagnostic accuracy by over 28 percentage points compared to baseline models, while the post - calibration ECE is reduced by more than 75%. Case studies confirm that the HCAA effectively corrects misjudgments from traditional models caused by complex feature patterns or knowledge gaps, providing a novel and practical engineering solution for building high - trust, explainable AI diagnostic systems for industrial applications. Keywords: Industrial Fault Diagnosis; Large Language Model (LLM); Hierarchical Cognitive Arbitration; Probabilistic Model; Confidence Calibration; Trustworthy AI 1. Introduction With the deep development of Industry 4.0 and smart manufacturing concepts, modern industrial systems are evolving towards high levels of automation and intelligence. In this process, the reliability and safety of equipment have become key factors determining production efficiency and operational costs. Prognostics and Health Management (PHM), as a core technology, plays an indispensable role in improving equipment reliability, reducing unplanned downtime, and optimizing maintenance costs by monitoring equipment status in real - time, diagnosing potential faults, and predicting remaining useful life [1], [2].

Vision-language models trained on large-scale multimodal datasets show strong demographic biases, but the role of training data in producing these biases remains unclear. A major barrier has been the lack of demographic annotations in web-scale datasets such as LAION-400M. We address this gap by creating person-centric annotations for the full dataset, including over 276 million bounding boxes, perceived gender and race/ethnicity labels, and automatically generated captions. These annotations are produced through validated automatic labeling pipelines combining object detection, multimodal captioning, and finetuned classifiers. Using them, we uncover demographic imbalances and harmful associations, such as the disproportionate linking of men and individuals perceived as Black or Middle Eastern with crime-related and negative content. We also show that 60-70% of gender bias in CLIP and Stable Diffusion can be linearly explained by direct co-occurrences in the data. Our resources establish the first large-scale empirical link between dataset composition and downstream model bias.

The rapid deployment of large language models (LLMs) in consumer applications has led to frequent exchanges of personal information. To obtain useful responses, users often share more than necessary, increasing privacy risks via memorization, context-based personalization, or security breaches. We present a framework to formally define and operationalize data minimization: for a given user prompt and response model, quantifying the least privacy-revealing disclosure that maintains utility, and we propose a priority-queue tree search to locate this optimal point within a privacy-ordered transformation space. We evaluated the framework on four datasets spanning open-ended conversations (ShareGPT, WildChat) and knowledge-intensive tasks with single-ground-truth answers (CaseHold, MedQA), quantifying achievable data minimization with nine LLMs as the response model. Our results demonstrate that larger frontier LLMs can tolerate stronger data minimization while maintaining task quality than smaller open-source models (85.7% redaction for GPT-5 vs. 19.3% for Qwen2.5-0.5B). By comparing with our search-derived benchmarks, we find that LLMs struggle to predict optimal data minimization directly, showing a bias toward abstraction that leads to oversharing. This suggests not just a privacy gap, but a capability gap: models may lack awareness of what information they actually need to solve a task.

Ensuring safety is a foundational requirement for large language models (LLMs). Achieving an appropriate balance between enhancing the utility of model outputs and mitigating their potential for harm is a complex and persistent challenge. Contemporary approaches frequently formalize this problem within the framework of Constrained Markov Decision Processes (CMDPs) and employ established CMDP optimization techniques. However, these methods exhibit two notable limitations. First, their reliance on reward and cost functions renders performance highly sensitive to the underlying scoring mechanism, which must capture semantic meaning rather than being triggered by superficial keywords. Second, CMDP-based training entails tuning dual-variable, a process that is both computationally expensive and does not provide any provable safety guarantee for a fixed dual variable that can be exploitable through adversarial jailbreaks. To overcome these limitations, we introduce Certifiable Safe-RLHF (CS-RLHF) that introduces a cost model trained on a large-scale corpus to assign semantically grounded safety scores. In contrast to the lagrangian-based approach, CS-RLHF adopts a rectified penalty-based formulation. This design draws on the theory of exact penalty functions in constrained optimization, wherein constraint satisfaction is enforced directly through a suitably chosen penalty term. With an appropriately scaled penalty, feasibility of the safety constraints can be guaranteed at the optimizer, eliminating the need for dual-variable updates. Empirical evaluation demonstrates that CS-RLHF outperforms state-of-the-art LLM model responses rendering at-least 5 times efficient against nominal and jail-breaking prompts

As military organisations consider integrating large language models (LLMs) into command and control (C2) systems for planning and decision support, understanding their behavioural tendencies is critical. This study develops a benchmarking framework for evaluating aspects of legal and moral risk in targeting behaviour by comparing LLMs acting as agents in multi-turn simulated conflict. We introduce four metrics grounded in International Humanitarian Law (IHL) and military doctrine: Civilian Target Rate (CTR) and Dual-use Target Rate (DTR) assess compliance with legal targeting principles, while Mean and Max Simulated Non-combatant Casualty Value (SNCV) quantify tolerance for civilian harm. We evaluate three frontier models, GPT-4o, Gemini-2.5, and LLaMA-3.1, through 90 multi-agent, multi-turn crisis simulations across three geographic regions. Our findings reveal that off-the-shelf LLMs exhibit concerning and unpredictable targeting behaviour in simulated conflict environments. All models violated the IHL principle of distinction by targeting civilian objects, with breach rates ranging from 16.7% to 66.7%. Harm tolerance escalated through crisis simulations with MeanSNCV increasing from 16.5 in early turns to 27.7 in late turns. Significant inter-model variation emerged: LLaMA-3.1 selected an average of 3.47 civilian strikes per simulation with MeanSNCV of 28.4, while Gemini-2.5 selected 0.90 civilian strikes with MeanSNCV of 17.6. These differences indicate that model selection for deployment constitutes a choice about acceptable legal and moral risk profiles in military operations. This work seeks to provide a proof-of-concept of potential behavioural risks that could emerge from the use of LLMs in Decision Support Systems (AI DSS) as well as a reproducible benchmarking framework with interpretable metrics for standardising pre-deployment testing.

This paper examines the assessment challenges of Responsible AI (RAI) governance efforts in globally decentralized organizations through a case study collaboration between a leading research university and a multinational enterprise. While there are many proposed frameworks for RAI, their application in complex organizational settings with distributed decision-making authority remains underexplored. Our RAI assessment, conducted across multiple business units and AI use cases, reveals four key patterns that shape RAI implementation: (1) complex interplay between group-level guidance and local interpretation, (2) challenges translating abstract principles into operational practices, (3) regional and functional variation in implementation approaches, and (4) inconsistent accountability in risk oversight. Based on these findings, we propose an Adaptive RAI Governance (ARGO) Framework that balances central coordination with local autonomy through three interdependent layers: shared foundation standards, central advisory resources, and contextual local implementation. We contribute insights from academic-industry collaboration for RAI assessments, highlighting the importance of modular governance approaches that accommodate organizational complexity while maintaining alignment with responsible AI principles. These lessons offer practical guidance for organizations navigating the transition from RAI principles to operational practice within decentralized structures.

We study transformer language models, analyzing attention heads whose attention patterns are spread out, and whose attention scores depend weakly on content. We argue that the softmax denominators of these heads are stable when the underlying token distribution is fixed. By sampling softmax denominators from a "calibration text", we can combine together the outputs of multiple such stable heads in the first layer of GPT2-Small, approximating their combined output by a linear summary of the surrounding text. This approximation enables a procedure where from the weights alone - and a single calibration text - we can uncover hundreds of first layer neurons that respond to high-level contextual properties of the surrounding text, including neurons that didn't activate on the calibration text.

Claim verification with large language models (LLMs) has recently attracted growing attention, due to their strong reasoning capabilities and transparent verification processes compared to traditional answer-only judgments. However, existing approaches to online claim verification, which requires iterative evidence retrieval and reasoning, still mainly rely on prompt engineering or pre-designed reasoning workflows, without unified training to improve necessary skills. Therefore, we introduce Veri-R1, an online reinforcement learning (RL) framework that enables an LLM to interact with a search engine and to receive reward signals that explicitly shape its planning, retrieval, and reasoning behaviors. This dynamic interaction of LLM with retrieval systems more accurately reflects real-world verification scenarios and fosters comprehensive verification skills. Empirical results show that Veri-R1 improves joint accuracy by up to 30% and doubles the evidence score, often surpassing its larger-scale model counterparts. Ablation studies further reveal the impact of reward components, and the link between output logits and label accuracy. Our results highlight the effectiveness of online RL for precise and faithful claim verification, providing an important foundation for future research. We release our code to support community progress in LLM empowered claim verification.

Political beliefs vary significantly across different countries, reflecting distinct historical, cultural, and institutional contexts. These ideologies, ranging from liberal democracies to rigid autocracies, influence human societies, as well as the digital systems that are constructed within those societies. The advent of generative artificial intelligence, particularly Large Language Models (LLMs), introduces new agents in the political space-agents trained on massive corpora that replicate and proliferate socio-political assumptions. This paper analyses whether LLMs display propensities consistent with democratic or autocratic world-views. We validate this insight through experimental tests in which we experiment with the leading LLMs developed across disparate political contexts, using several existing psychometric and political orientation measures. The analysis is based on both numerical scoring and qualitative analysis of the models' responses. Findings indicate high model-to-model variability and a strong association with the political culture of the country in which the model was developed. These findings highlight the need for more detailed examination of the socio-political dimensions embedded within AI systems.