AI Catalogue vocabulary

AI Catalogue vocabulary (AICat) is a semantic-based vocabulary designed to facilitate the documentation, risk classification, and registration of AI systems. Built using RDF and RTO technologies, it enables AI systems to be described in a machine-readable format. This addresses the EU AI Act's requirement for standardized technical documentation (Article 48) and risk-adjusted obligations (Article 6). Unlike static documents, AICat allows for automated compliance checks, ensuring that AI systems can be queried and verified by both regulators and internal auditors. It aligns with ISO 42001 AI Management System standards, providing a common language for AI risk-adjusted governance across different organizational-level AI deployments.

Implementation typically follows three steps: (1) Semantic Mapping: Converting existing AI technical documentation into AICat-compliant RDF format. (2) Automated Risk Classification: Using SPARQL queries to automatically categorize AI systems according to EU AI Act risk tiers (Unacceptable, High, Limited, Minimal). (3) Continuous Monitoring: Updating the AI catalogue as systems evolve. For example, a European fintech firm using AICat for its credit scoring AI could automatically detect when a model update changes its risk profile from 'Limited' to 'High', triggering a mandatory human oversight review. This proactive approach can reduce compliance-related delays by up to 50% compared to manual processes.

Taiwan enterprises face three primary challenges: (1) Technical Expertise: Semantic Web technologies like RDF/SPARQL are niche skills in the local AI market, requiring upskilling or external expertise. (2) Data Silos: AI systems are often managed by different departments (IT, HR, Legal) with no unified data-sharing mechanism, necessitating a centralized AI catalogue. (3) Regulatory Uncertainty: As Taiwan' AI basic law is still in development, companies may be unsure which standards to prioritize. The solution is to adopt international standards like ISO 42001 and EU AI Act as the baseline, ensuring the AI catalogue remains relevant even as local regulations evolve. Companies should start with a pilot of 3-5 high-impact AI systems before scaling enterprise-wide.

Winners Consulting Services Co., Ltd. specializes in AI Catalogue vocabulary for Taiwan enterprises, delivering compliant management systems within 90 days. Free consultation: https://winners.com.tw/contact