Technology Category
Philosophy and Ethics of AI: Safety, Robustness & TrustworthinessData Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, TrustHumans and AI: Human-Aware Planning and Behavior Prediction
Towards a Framework for Operationalizing the Specification of Trustworthy AI Requirements
🤖 AI Summary
Emerging, context-dependent trustworthiness requirements in AI systems lack structured specification methods. Method: This paper proposes the first integrated trustworthy AI requirements framework unifying AMDiRE (Artifact-based Requirements Engineering) and PerSpecML (multi-perspective specification for ML systems), enabling systematic modeling, traceability, and verification of trustworthiness requirements for non-deterministic AI operating in dynamic environments. Contribution/Results: It introduces the first deep integration of artifact-based requirements engineering with multi-perspective ML specification, establishing an end-to-end mapping from abstract stakeholder trust concerns to executable, formal requirements specifications. A prototype implementation demonstrates feasibility and identifies key open challenges—scalability, cross-perspective consistency, and runtime adaptability—thereby providing a systematic foundation for trustworthy AI requirements engineering practice.
Technology Category
Application Category
📝 Abstract
Growing concerns around the trustworthiness of AI-enabled systems highlight the role of requirements engineering (RE) in addressing emergent, context-dependent properties that are difficult to specify without structured approaches. In this short vision paper, we propose the integration of two complementary approaches: AMDiRE, an artefact-based approach for RE, and PerSpecML, a perspective-based method designed to support the elicitation, analysis, and specification of machine learning (ML)-enabled systems. AMDiRE provides a structured, artefact-centric, process-agnostic methodology and templates that promote consistency and traceability in the results; however, it is primarily oriented toward deterministic systems. PerSpecML, in turn, introduces multi-perspective guidance to uncover concerns arising from the data-driven and non-deterministic behavior of ML-enabled systems. We envision a pathway to operationalize trustworthiness-related requirements, bridging stakeholder-driven concerns and structured artefact models. We conclude by outlining key research directions and open challenges to be discussed with the RE community.
Problem
Research questions and friction points this paper is trying to address.
Structured approach for specifying trustworthy AI requirements
Bridging stakeholder concerns with artifact models
Addressing non-deterministic behavior in ML systems
Innovation
Methods, ideas, or system contributions that make the work stand out.
Integrates AMDiRE and PerSpecML for trustworthy AI
AMDiRE offers structured, artefact-centric RE methodology
PerSpecML provides multi-perspective ML system guidance
🔎 Similar Papers
No similar papers found.
