​INTERNSHIP APPLICATIONS

• Issue: Lack of Explainability. The school “cannot explain the AI’s decision logic” for rejecting candidates. This directly matches “Lack of Explainability,” meaning “it is not possible to explain how the AI system works or why it made a certain decision”.

• Safeguard: Explainable AI. Use or require an AI system that provides clear, understandable reasons for each decision. This ensures the school could give feedback to rejected students and verify fairness.

• Low: While unfair explanations are problematic, the harm (frustrated applicants) is limited compared to other scenarios.
   

• Decision: Accept. The risk is relatively low and inherent to using this opaque AI. We would accept the risk but try to mitigate its effects by seeking tools with better explainability or by manually providing context to students.