Assessing AI – a View on AI Conformity and Quality in Europe

As Europe moves towards the implementation of the AI Act, the role of third-party assessment and certification will become a cornerstone of safe and trustworthy AI. From mid-2027 onwards, conformity assessment for high-risk AI systems will be mandatory, placing significant responsibility on independent actors to evaluate compliance.

In this talk, I will share insights from TÜV AI.Lab, a company at the forefront of developing certification processes for AI systems across sectors such as healthcare, automotive, and industrial machinery. I will highlight the unique challenges with regard to testing and certifying AI systems, scaling AI assessment methods and assessing conformity with the AI Act.

I will also reflect on the point that thorough assessment and testing not only ensure compliance, but also help companies deliver higher-quality products, improve client satisfaction, and reduce reputational and operational risks.

Lastly, I will briefly touch upon what we see as necessary steps for a smooth implementation of the AI Act in Germany and Europe.