powered by KI.NRW: Inspection catalog for AI systems published

For more quality, security and competitiveness: Guidelines help making AI systems more trustworthy

From voice assistance systems to the analysis of job application documents and autonomous driving – artificial intelligence (AI) is already being used in many areas of everyday life. This means it is even more important to design AI applications so that they act reliably and securely and handle data transparently and trustworthily. Supported by the Competence Platform KI.NRW, scientists from the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS have now published an AI inspection catalog of over 160 pages, which is designed to help make intelligent applications more trustworthy. As a guidance, it supports developers in the design and testers in the evaluation and quality assurance of AI systems.

“With the AI inspection catalog, we now have a practical document that can be used by independent testing organizations as a basis for future product testing. At the same time, it provides companies with the tools they need to evaluate and improve their systems themselves already during the development process and thus be prepared for future regulatory requirements” says Prof. Dr. Stefan Wrobel, Institute Director Fraunhofer IAIS and member of the AI Standardization and Conformity Coordination Group newly established by the German government.

For the competitiveness of companies, testing and certification of AI systems is quite central: “The inspection catalog offers a concrete guideline with which companies can develop their AI systems to high quality standards. This is important to ensure trust in our technologies and to succeed on international markets” says NRW Economics Minister Prof. Dr. Andreas Pinkwart, who was handed the first version today.