What issue arises when AI creates unfair outcomes due to biased training data?

The issue that arises when AI creates unfair outcomes due to biased training data is best characterized as algorithmic bias. This occurs when the algorithms used in AI systems produce systematic and unfair discrimination against certain groups or individuals, often reflecting stereotypes or prejudices that were present in the training data.

When training datasets contain biases—whether they are based on gender, race, socio-economic status, or other factors—AI systems can learn these biases and perpetuate them in their outputs, leading to unfair treatment in scenarios such as hiring practices, loan approvals, law enforcement, and many other applications.

Addressing algorithmic bias is crucial for developing ethical AI systems, ensuring they operate fairly and justly in real-world applications. This focuses on the need for rigorous scrutiny of the data used for training and implementing measures to mitigate biases in AI outputs.