Clients are concerned that generative AI models will propagate established biases and hate, or unethical behavior, which could be very damaging to their company, not to mention have legal ramifications. IBM knows that AI models must be architected with governance in mind from the start, not as an afterthought. For responsible AI, the objective is to provide transparency and explainability for models and to support increasing regulatory compliance demands. What is the IBM core principle that underpins this?