Towards a Framework for Supporting the Ethical and Regulatory Certification of AI Systems