An AI system has been found to discriminate against members of particular federally protected classes. What is the corresponding ethical concern to the issue described in the scenario?

Safety pertains to the protection of individuals from harm or danger, often in physical or operational contexts. While safety is important in AI deployment, the scenario specifically addresses discriminatory practices, which are more directly related to fairness rather than safety concerns. Thus, safety does not capture the ethical implications of biased AI behavior.

Transparency involves clear communication about how AI systems operate, including their decision-making processes and underlying algorithms. Although transparency is essential for understanding AI biases, it does not directly address the ethical concern of discrimination itself. The core issue here is not how decisions are made but the fairness of those decisions toward protected classes.

Privacy concerns focus on the protection of personal information and individuals' rights to control their data. While privacy is a vital ethical issue in AI, the scenario centers on discriminatory outcomes rather than the handling of personal data. Thus, privacy does not appropriately represent the ethical dilemma posed by the discrimination described.