Romain Bourboulou

In today's rapidly evolving AI landscape, thorough testing of AI systems has become critical yet significantly more complex than traditional software testing. Organizations struggle with selecting appropriate evaluation methods for different AI modalities, interpreting benchmark results accurately, and establishing reliable validation processes that address fairness, safety, and performance concerns.

·       Implementing testing strategies tailored to specific AI types and use cases

·       Evaluating benchmark reliability through structured criteria and statistical validation

·       Designing comprehensive testing frameworks that anticipate regulatory requirements

This interactive workshop provides practical experience with diverse AI testing methodologies across different system types. You'll depart with a customizable testing framework, techniques to critically assess benchmark results, and strategies to build robust validation processes that can adapt to emerging AI capabilities and governance requirements.