Explainable AI models

Ensuring that the AI models we use are explainable and ethical so that we can make sure they are robust and reliable, but also safe, secure, transparent, fair, inclusive, and produce accountable results.

1.- Develop and explain our own AI models Quality (of data) inspector. Support from model to code. Explainable AI to reduce the cost of errors. Scenarios coverage and representativeness. ML and generation of synthetic data to support calibration (automotive).

2.- Explain existing AI models Identify which AI application needs to be explained. Build explainability model. Design visualisation to translate XAI for appropriate stakeholders (also non-technical decision makers, through GenAI agents).