Image de couverture de livre

Safety is Non-Compositional: A Formal Framework for Capability-Based AI Systems

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteur principal:	Spera, Cosimo
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

From Workflow Automation to Capability Closure: A Formal Framework for Safe and Revenue-Aware Customer Service AI
par: Spera, Cosimo
Publié: (2026)

Capability Safety as Datalog: A Foundational Equivalence
par: Spera, Cosimo
Publié: (2026)

Safety as Computation: Certified Answer Reuse via Capability Closure in Task-Oriented Dialogue
par: Spera, Cosimo
Publié: (2026)

Reversing the Paradigm: Building AI-First Systems with Human Guidance
par: Spera, Cosimo, et autres
Publié: (2025)

Redefining CX with Agentic AI: Minerva CQ Case Study
par: Agrawal, Garima, et autres
Publié: (2025)

Beyond-RAG: Question Identification and Answer Generation in Real-Time Conversations
par: Agrawal, Garima, et autres
Publié: (2024)

BlueGlass: A Framework for Composite AI Safety
par: Nandigramwar, Harshal, et autres
Publié: (2025)

Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems
par: Allegrini, Edoardo, et autres
Publié: (2025)

Safe Bilevel Delegation (SBD): A Formal Framework for Runtime Delegation Safety in Multi-Agent Systems
par: Sun, Yuan
Publié: (2026)

A Conceptual Framework for AI Capability Evaluations
par: Carro, María Victoria, et autres
Publié: (2025)

With Great Capabilities Come Great Responsibilities: Introducing the Agentic Risk & Capability Framework for Governing Agentic AI Systems
par: Khoo, Shaun, et autres
Publié: (2025)

Evaluating Human-AI Safety: A Framework for Measuring Harmful Capability Uplift
par: Vaccaro, Michelle, et autres
Publié: (2026)

HySafe-AI: Hybrid Safety Architectural Analysis Framework for AI Systems: A Case Study
par: Pitale, Mandar, et autres
Publié: (2025)

Using Formal Models, Safety Shields and Certified Control to Validate AI-Based Train Systems
par: Gruteser, Jan, et autres
Publié: (2024)

A Framework for Evaluating Emerging Cyberattack Capabilities of AI
par: Rodriguez, Mikel, et autres
Publié: (2025)

SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of Foundation Model-based Embodied Agents
par: Zhan, Simon Sinong, et autres
Publié: (2025)

The AI Pyramid A Conceptual Framework for Workforce Capability in the Age of AI
par: Khatri, Alok, et autres
Publié: (2026)

DeepKnown-Guard: A Proprietary Model-Based Safety Response Framework for AI Agents
par: Li, Qi, et autres
Publié: (2025)

Let's Reason Formally: Natural-Formal Hybrid Reasoning Enhances LLM's Math Capability
par: Wang, Ruida, et autres
Publié: (2025)

Defining Operational Conditions for Safety-Critical AI-Based Systems from Data
par: Christensen, Johann Maximilian, et autres
Publié: (2026)

OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety
par: Vijayvargiya, Sanidhya, et autres
Publié: (2025)

AI Safety is Stuck in Technical Terms -- A System Safety Response to the International AI Safety Report
par: Dobbe, Roel
Publié: (2025)

Semantic Entanglement in Vector-Based Retrieval: A Formal Framework and Context-Conditioned Disentanglement Pipeline for Agentic RAG Systems
par: Loghmani, Nick
Publié: (2026)

AI Safety as Control of Irreversibility: A Systems Framework for Decision-Energy and Sovereignty Boundaries
par: Shu, Wesley, et autres
Publié: (2026)

Domain-Agnostic Scalable AI Safety Ensuring Framework
par: Kim, Beomjun, et autres
Publié: (2025)

X-RAY: Mapping LLM Reasoning Capability via Formalized and Calibrated Probes
par: Tianxi, Gao, et autres
Publié: (2026)

SafeEmbodAI: a Safety Framework for Mobile Robots in Embodied AI Systems
par: Zhang, Wenxiao, et autres
Publié: (2024)

PACEbench: A Framework for Evaluating Practical AI Cyber-Exploitation Capabilities
par: Liu, Zicheng, et autres
Publié: (2025)

A Formal Security Framework for MCP-Based AI Agents: Threat Taxonomy, Verification Models, and Defense Mechanisms
par: Acharya, Nirajan, et autres
Publié: (2026)

FormalGeo: An Extensible Formalized Framework for Olympiad Geometric Problem Solving
par: Zhang, Xiaokai, et autres
Publié: (2023)

Safety Cases: How to Justify the Safety of Advanced AI Systems
par: Clymer, Joshua, et autres
Publié: (2024)

Emerging Practices in Frontier AI Safety Frameworks
par: Buhl, Marie Davidsen, et autres
Publié: (2025)

A Defect Classification Framework for AI-Based Software Systems (AI-ODC)
par: Alannsary, Mohammed O.
Publié: (2025)

Quantifying Multimodal Capabilities: Formal Generalization Guarantees in Pairwise Metric Learning
par: Zhou, Richeng, et autres
Publié: (2026)

FVEval: Understanding Language Model Capabilities in Formal Verification of Digital Hardware
par: Kang, Minwoo, et autres
Publié: (2024)

Fighting AI with AI: Leveraging Foundation Models for Assuring AI-Enabled Safety-Critical Systems
par: Mavridou, Anastasia, et autres
Publié: (2025)

The Institutional Scaling Law: Non-Monotonic Fitness, Capability-Trust Divergence, and Symbiogenetic Scaling in Generative AI
par: Baciak, Mark, et autres
Publié: (2026)

A Dynamical Systems Framework for Reinforcement Learning Safety and Robustness Verification
par: Nasir, Ahmed, et autres
Publié: (2025)

AI-Educational Development Loop (AI-EDL): A Conceptual Framework to Bridge AI Capabilities with Classical Educational Theories
par: Yu, Ning, et autres
Publié: (2025)

Investigating Language Model Capabilities to Represent and Process Formal Knowledge: A Preliminary Study to Assist Ontology Engineering
par: Akl, Hanna Abi
Publié: (2025)