This session explores the critical issues surrounding bias, accountability, and transparency in AI systems. Topics include fairness-aware algorithms, explainable ML models (e.g., LIME, SHAP), and methods to detect and mitigate bias in training data. Regulatory frameworks and ethical guidelines will be examined, along with real-world failures and lessons learned. Participants will be encouraged to reflect on the societal impacts of AI, particularly in sensitive domains such as criminal justice, hiring, and lending. The session will include discussions on fostering inclusive datasets and participatory design methodologies.