Technical reports · 2019
Trustworthy Use of Artificial Intelligence
Vertrauenswürdiger Einsatz von Künstlicher Intelligenz
Overview
Examines technical reliability alongside ethical and legal perspectives through six assessment areas: autonomy and control, fairness, transparency, reliability, safety/security and data protection. Discusses training-data representativeness, model uncertainty and system integration as issues for an emerging AI assessment framework.
Publication and version information
A 2019 whitepaper discussing concepts and priorities for a subsequent AI assessment catalogue.
Version: 2019 whitepaper; publisher German and English versions