With the development of Large Language Models (LLMs), numerous efforts have revealed their vulnerabilities to jailbreak attacks. Although these studies have driven the progress in LLMs' safety alignment, it remains unclear whether LLMs have internalized authentic knowledge to deal with real-world crimes, or are merely forced to simulate toxic language patterns. This ambiguity raises concerns that jailbreak success is often attributable to a hallucination loop between jailbroken LLM and judger LLM. By decoupling the use of jailbreak techniques, we construct knowledge-intensive Q\&A to investigate the misuse threats of LLMs in terms of dangerous knowledge possession, harmful task planning utility, and harmfulness judgment robustness. Experiments reveal a mismatch between jailbreak success rates and harmful knowledge possession in LLMs, and existing LLM-as-a-judge frameworks tend to anchor harmfulness judgments on toxic language patterns. Our study reveals a gap between existing LLM safety assessments and real-world threat potential.

We propose an open-source framework for Command-style dictation that addresses the gap between resource-intensive Online systems and high-latency Batch processing. Our approach uses Voice Activity Detection (VAD) to segment audio and transcribes these segments in parallel using Whisper models, enabling efficient multiplexing across audios. Unlike proprietary systems like SuperWhisper, this framework is also compatible with most ASR architectures, including widely used CTC-based models. Our multiplexing technique maximizes compute utilization in real-world settings, as demonstrated by its deployment in around 15% of India's courtrooms. Evaluations on live data show consistent latency reduction as user concurrency increases, compared to sequential batch processing. The live demonstration will showcase our open-sourced implementation and allow attendees to interact with it in real-time.

PEER-REVIEWED AND ACCEPTED IN IEEE- ISTAS 2025 The rapid evolution of Generative AI (GenAI) has introduced unprecedented opportunities while presenting complex challenges around ethics, accountability, and societal impact. This paper draws on a literature review, established governance frameworks, and industry roundtable discussions to identify core principles for integrating responsible GenAI governance into diverse organizational structures. Our objective is to provide actionable recommendations for a balanced, risk-based governance approach that enables both innovation and oversight. Findings emphasize the need for adaptable risk assessment tools, continuous monitoring practices, and cross-sector collaboration to establish trustworthy GenAI. These insights provide a structured foundation and Responsible GenAI Guide (ResAI) for organizations to align GenAI initiatives with ethical, legal, and operational best practices.

Multi-agent large language model simulations have the potential to model complex human behaviors and interactions. If the mechanics are set up properly, unanticipated and valuable social dynamics can surface. However, it is challenging to consistently enforce simulation mechanics while still allowing for notable and emergent dynamics. We present AgentDynEx, an AI system that helps set up simulations from user-specified mechanics and dynamics. AgentDynEx uses LLMs to guide users through a Configuration Matrix to identify core mechanics and define milestones to track dynamics. It also introduces a method called \textit{nudging}, where the system dynamically reflects on simulation progress and gently intervenes if it begins to deviate from intended outcomes. A technical evaluation found that nudging enables simulations to have more complex mechanics and maintain its notable dynamics compared to simulations without nudging. We discuss the importance of nudging as a technique for balancing mechanics and dynamics of multi-agent simulations.

Membership in ACM includes a subscription to Communications of the ACM (CACM), the computing industry's most trusted source for staying connected to the world of advanced computing. Interconnected systems have become the backbone of modern societies. However, the very same critical role played by these systems brings significant challenges: Securing interconnected systems is not merely a technological necessity, but a cornerstone for safeguarding the economic, political, and social stability of countries. While these challenges are global, the Arab World presents a unique landscape that warrants a nuanced exploration of both commonalities and peculiarities within the broader context of securing interconnected systems (see Figure for a brief summary of these challenges). Interconnected systems, including cyber-physical systems, often combine computational and physical processes. They include critical infrastructure such as power grids, transportation networks, and healthcare systems, alongside commercial and industrial applications.

Over 200 contractors who work on improving Google's AI products, including Gemini and AI Overviews, have been laid off, sources say. Workers enter a building on the Google headquarters campus on July 23, 2025, in Mountain View, California. More than 200 contractors who worked on evaluating and improving Google's AI products have been laid off without warning in at least two rounds of layoffs last month. The move comes amid an ongoing fight over pay and working conditions, according to workers who spoke to WIRED. In the past few years, Google has outsourced its AI rating work--which includes evaluating, editing, or rewriting the Gemini chatbot's response to make it sound more human and "intelligent"--to thousands of contractors employed by Hitachi-owned GlobalLogic and other outsourcing companies.

It's Not Going to Get It From Trump's Corporate-Friendly EPA The White House's new Make America Healthy Again strategy makes some asks of the EPA--but critics say the agency is too industry-friendly to make a difference. When Jean-Marie Kauth first read the Make America Healthy Again commission report, released by the White House in May, she was "thrilled about some of the things they identified," she says. "They clearly called out industry as a pernicious influence on why EPA has not been very successful in regulating chemicals, especially pesticides." Kauth's daughter died of leukemia at age 8 after, Kauth says, she was exposed to the insecticide chlorpyrifos, which the EPA banned in 2021. Kauth, a professor at Benedictine University in Illinois, now serves as a member of the EPA's Children's Health Protection Advisory Committee (CHPAC), a group of outside experts who advise the agency on children's health issues.

The absolute dominance of Artificial Intelligence (AI) introduces unprecedented societal harms and risks. Existing AI risk assessment models focus on internal compliance, often neglecting diverse stakeholder perspectives and real-world consequences. We propose a paradigm shift to a human-centric, harm-severity adaptive approach grounded in empirical incident data. We present AI Harmonics, which includes a novel AI harm assessment metric (AIH) that leverages ordinal severity data to capture relative impact without requiring precise numerical estimates. AI Harmonics combines a robust, generalized methodology with a data-driven, stakeholder-aware framework for exploring and prioritizing AI harms. Experiments on annotated incident data confirm that political and physical harms exhibit the highest concentration and thus warrant urgent mitigation: political harms erode public trust, while physical harms pose serious, even life-threatening risks, underscoring the real-world relevance of our approach. Finally, we demonstrate that AI Harmonics consistently identifies uneven harm distributions, enabling policymakers and organizations to target their mitigation efforts effectively.

Large Language Models (LLMs) have significantly advanced the development of Legal Artificial Intelligence (Legal AI) in recent years, enhancing the efficiency and accuracy of legal tasks. To advance research and applications of LLM-based approaches in legal domain, this paper provides a comprehensive review of 16 legal LLMs series and 47 LLM-based frameworks for legal tasks, and also gather 15 benchmarks and 29 datasets to evaluate different legal capabilities. Additionally, we analyse the challenges and discuss future directions for LLM-based approaches in the legal domain. We hope this paper provides a systematic introduction for beginners and encourages future research in this field. Resources are available at https://github.com/ZhitianHou/LLMs4LegalAI.

The practical adoption of deep learning in high-stakes forensic applications, such as dental age estimation, is often limited by the 'black box' nature of the models. This study introduces a framework designed to enhance both performance and transparency in this context. We use a notable performance disparity in the automated staging of mandibular second (tooth 37) and third (tooth 38) molars as a case study. The proposed framework, which combines a convolutional autoencoder (AE) with a Vision Transformer (ViT), improves classification accuracy for both teeth over a baseline ViT, increasing from 0.712 to 0.815 for tooth 37 and from 0.462 to 0.543 for tooth 38. Beyond improving performance, the framework provides multi-faceted diagnostic insights. Analysis of the AE's latent space metrics and image reconstructions indicates that the remaining performance gap is data-centric, suggesting high intra-class morphological variability in the tooth 38 dataset is a primary limiting factor. This work highlights the insufficiency of relying on a single mode of interpretability, such as attention maps, which can appear anatomically plausible yet fail to identify underlying data issues. By offering a methodology that both enhances accuracy and provides evidence for why a model may be uncertain, this framework serves as a more robust tool to support expert decision-making in forensic age estimation.