AI Ethics

Unpacking the black box: Transparency AI implementations are often black boxes. A black box in AI is when we know the inputs and the outputs. Ambiguousness is non-ideal Question of trust: Can we validate AI decisions without understanding them? Transparency: Making an AI’s decision-making process understandable. Throughout the AI life cycle Transparency in AI involves in all stages of AI life cycle. Data Collection - Data Preparation - Model Training - Model Evaluation - Model Deployment Purpose: Understanding the workings of the AI system. Getting comfortable with the operation of AI. Transparency in AI can be intimidating but is beneficial for businesses. Transparency leads to predictable regulations and public preception.

AI Fairness Ensure no group is favored over another. concerns race, gender, socioeconomic status etc. AI should predict outcomes equitably. There should be no bias towards any group.