arXiv:2607. 23365v1 Announce Type: cross Abstract: Artificial intelligence (AI) systems are increasingly deployed in high-stakes domains such as healthcare, autonomous driving, finance, and education.
arXiv:2608. 07446v1 Announce Type: cross Abstract: Rapid adoption of large language models (LLMs) in enterprise settings has introduced operational, security, and governance risks.
The paper proposes AI Deployment Accountability Engineering (ADAE), a new subdiscipline focused on establishing measurable, continuous, and actionable accountability for AI systems once they are deployed. ADAE treats accountability as a deployment-layer property, aiming to ensure systems remain within acceptable risk limits, identify failure contexts, attribute failures across technical and human components, and translate technical failures into downstream consequences. The authors outline a research agenda built around four pillars—structured discovery of context-dependent failure modes, privacy-preserving accountability measurement, system-level risk analysis for agentic AI, and translation of technical failures into operational and institutional risks—to support timely intervention in safety-critical socio-technical environments.
The paper systematically classifies the EU AI Act’s high‑risk requirements, finding that only a minority directly address AI‑specific risk sources while most impose organizational and documentation obligations. From these risk‑related requirements, the authors derive a consolidated list of distinct AI‑specific risk sources, creating an EU AI Act Risk Source List. This list aims to bridge the gap between legal obligations and AI risk‑management practice by providing a structured reference for comparing the Act’s implicit risk coverage with existing AI risk taxonomies.
arXiv:2608. 02638v1 Announce Type: cross Abstract: Artificial Intelligence (AI) components are increasingly pervasive in several software systems, including Cyber-Physical Systems (CPSs).
arXiv:2607. 16130v1 Announce Type: cross Abstract: AI governance increasingly requires judgments about whether an AI system remains adequately trustworthy over time, whether observed changes are tolerable, and how such judgments should be documented in a transparent and contestable way.
The paper "Operationalising AI Regulatory Sandboxes: Activities, Requirements, and Technical Assessment under the EU AI Act" outlines a detailed framework for implementing AI Regulatory Sandboxes (AIRS) under the EU AI Act. It maps the sandbox lifecycle into 29 activities, distinguishes between a Core AIRS and an Extended AIRS that includes an AI Technical Sandbox (AITS), and derives 15 infrastructural and governance requirements linked to these activities and provider obligations. The authors also introduce the Sandbox Configurator, an open‑source tool to instantiate AITS environments, aiming to provide structured workflows for regulators, robust evaluation methods for experts, and a transparent compliance pathway for AI providers.
arXiv:2607. 19292v1 Announce Type: cross Abstract: Current AI safety discourse still focuses disproportionately on visible failures, including obvious harms, dramatic misuse, and hypothetical catastrophic scenarios.
The paper proposes rethinking bias in AI as a diagnostic tool rather than merely a flaw to be minimized. It introduces a multidimensional framework that examines bias across origin, lifecycle emergence, technical causes, and validation methods, covering 30 bias types, 16 verification methods, and 20 countermeasures for both traditional and generative AI. The authors present a hierarchical evidence framework distinguishing internal and external validity, and advocate for Ethics by Design principles to embed bias verification throughout the AI development lifecycle.
arXiv:2607. 02197v1 Announce Type: cross Abstract: The society and emerging risk-based regulatory frameworks for AI underscore the need for rigorous risk assessment to ensure safe and reliable AI systems.
arXiv:2608. 04921v1 Announce Type: cross Abstract: As AI systems become increasingly integrated into diverse interfaces and applications, model-centric audits are insufficient to address risks arising from interactions among system components and deployment environments.
The paper introduces the concept of Evolutionary Safety for recursive self-improving AI, focusing on how safety properties evolve as an AI system and its successors change. It identifies key risks such as intent drift, error accumulation, and safety-property erosion, and presents a taxonomy covering agent state, model state, evaluation, environment, and update mechanisms. The authors propose methods for discovering and evaluating evolutionary risks, and outline governance principles for modification, selection, authorization, provenance, and recovery, while highlighting open problems for maintaining safety in persistent, adaptive, and recursively self-improving systems.
arXiv:2410. 22526v2 Announce Type: replace Abstract: To effectively address potential harms from Artificial Intelligence (AI) systems, it is essential to identify and mitigate system-level hazards.