FairnessДЇairness is a core ethical principle that all humans aim to understand and apply. Use explainers to understand what mainly influences the model and domain experts to validate these influences.Д®xplore model interpretability in Azure Machine Learning to learn more about explainability. Explainers like LIME or SHapley Additive exPlanations (SHAP) interpret these models by analyzing the relationship between the input and output.ДЇairlearn is an Azure Machine Learning integration and an open-source toolkit for the SDK and the AutoML graphical user interface. For these models, use Explainable Boosting Machine, which is the state of the algorithm based on a decision tree or linear models, provides lossless explanations, and is editable by domain experts.Д«lack-box models are more challenging to interpret because of a complex internal structure, the neural network. Glass-box models are interpretable because of their structure. Microsoft has developed InterpretML, an open-source toolkit that helps to achieve model explainability and supports glass-box and black-box models. Likewise, in order to comply with the company's policies, an auditor needs a tool that validates the model, and a business decision maker needs to be able to provide a transparent model in order to gain trust. A data scientist should be able to explain to the stakeholder how they achieved certain levels of accuracy and what influenced this outcome. This also ensures compliance with company policies, industry standards, and government regulations. ExplainabilityД®xplainability helps data scientists, auditors, and business decision makers to ensure that AI systems can reasonably justify their decisions and how they reach their conclusions. Rigorous testing and validation should be established for operating conditions to ensure that the system responds safely to edge cases, and A/B testing and champion/challenger methods should be integrated into the evaluation process.ДЄn AI system's performance can degrade over time, so a robust monitoring and model tracking process needs to be established to reactively and proactively measure the model's performance and retrain it, as necessary, to modernize it. Its inherent resilience should resist intended or unintended manipulation. It's important for a system to perform as it was originally designed and for it to respond safely to new situations. Reliability and safetyДЄI systems need to be reliable and safe in order to be trusted. Where possible, speech-to-text, text-to-speech, and visual recognition technology should be used to empower people with hearing, visual, and other impairments. Inclusiveness mandates that AI should consider all human races and experiences, and inclusive design practices can help developers to understand and address potential barriers that could unintentionally exclude people. While this guidance might vary depending on the company and region, it should reflect an organization's AI journey. Organizations should consider establishing an internal review body that provides oversight, insights, and guidance about developing and deploying AI systems.

The people who design and deploy the AI system need to be accountable for its actions and decisions, especially as we progress toward more autonomous systems. AccountabilityДЄccountability is an essential pillar of responsible AI. Learn more about Aether at this Microsoft Learn module. The core responsibility of the committee is to advise on responsible issues, technologies, processes, and best practices.

Microsoft established an ethical committee for AI, ethics, and effects in engineering and research Aether, in 2017. They are guided by two perspectives: ethical and explainable.ДЇrom an ethical perspective, AI should be fair and inclusive in its assertions, be accountable for its decisions, and not discriminate or hinder different races, disabilities, or backgrounds. These principles are essential to creating responsible and trustworthy AI as it moves into more mainstream products and services. Microsoft outlines six key principles for responsible AI: accountability, inclusiveness, reliability and safety, fairness, transparency, and privacy and security.

Microsoft toolkit 2.4.5 trusted