Evaluation, Ethics and Responsible AI

SIMFONI develops and applies robust evaluation, ethical oversight and governance frameworks to support the safe and responsible use of foundation models in healthcare.

Our work focuses on ensuring that AI systems are clinically safe, trustworthy and fit for real-world use. This includes:

• Developing evaluation frameworks for large language and multimodal models

• Strengthening model transparency and interpretability

• Implementing safeguards for clinical and technical safety

• Overseeing data governance and responsible data use

This work is carried out in collaboration with academic and ethics partners and in close engagement with regulators.