Authors: J. Kelly, S. Ali Zafar, L. Heidemann, J. Zacchi, D. Espinoza, N. Mata

Published on: March 25, 2024

Impact Score: 7.6

Arxiv code: Arxiv:2403.16808

Summary

• What is new: Proposes an extended product quality model for AI systems tailored to the EU AI Act for high-risk cases.
• Why this is important: Existing product quality models for AI do not adequately address the new requirements of the EU AI Act.
• What the research proposes: An extended product quality model mapping AI Act requirements to relevant quality attributes, introducing a contract-based approach for stakeholder-specific technical requirements.
• Results: Demonstrated applicability in an automotive supply chain scenario, ensuring AI systems compliance with EU AI Act’s safety and legality aspects.

Technical Details

Technological frameworks used: Extended product quality model for AI, Contract-based approach

Models used: Not specified

Data used: EU AI Act, Automotive supply chain case study

Potential Impact

Automotive supply chains, other high-risk AI application industries (healthcare, finance), AI development and regulation consultancy firms

Want to implement this idea in a business?

We have generated a startup concept here: ComplAI.