Description of the service

We audit training, validation, test and enterprise datasets for AI systems, especially for manufacturing use cases such as visual quality inspection, predictive maintenance, robotics and process optimization. The audit assesses whether the data is correct, complete, relevant and representative, including edge cases and relevant subgroups. We also review GDPR aspects, data origin, intellectual property risks and data management processes. The service supports preparation for the EU AI Act, especially Article 10 on data and data governance, and is aligned with ISO/IEC 5259, ISO 8000 and GDPR requirements.
Expected results:

The customer receives a structured audit report with a rating per criterion, identified weaknesses and prioritized recommendations. For a Deep Audit, the report also includes statistical analyses and quantified findings.

The methodology:

The service is available as an Assessment or as a Deep Audit. In the Assessment, the customer completes a structured questionnaire and we assess the answers for plausibility and completeness. This is suitable for an initial status review, internal preparation or early conformity planning and usually takes 2–4 weeks. In the Deep Audit, we receive access to the dataset and perform independent technical and statistical analyses. This is suitable for high-risk AI systems, external documentation or evidence towards customers, partners or authorities and usually takes 4–8 weeks.

Target:

Open for anyone

Improve your project
around production and
automation with AI technologies