AI Bias refers to unfair or prejudiced outcomes produced by an AI system, usually due to biased training data or flawed algorithm design.
Example: In the case study, a healthcare algorithm was trained on healthcare expense data instead of actual illness data. Since less money was spent on patients from the western region, the algorithm wrongly rated them as lower risk, even when they were more severely ill — showing clear bias against a particular group.
AI Access refers to unequal availability of AI tools, benefits, or technologies to different sections of society.
Example: If an advanced AI-based medical diagnosis tool is only available in urban hospitals and not in rural areas, people in remote regions are denied equal access to AI-driven healthcare benefits.
---