An AI assistant deployed for internal employee use begins producing outputs that reference confidential merger discussions after being connected to the company's document management system. What type of AI risk does this represent?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — when your AI assistant starts pulling confidential merger docs from the document management system, that's data leakage through the retrieval system. The access controls let information cross boundaries it shouldn't.
Full Explanation
The data leakage option is correct because the assistant retrieves confidential merger documents from the document management system, showing retrieval access controls are insufficient and confidential information crosses its intended access boundary. The hallucination option is wrong because it requires evidence the merger details were fabricated rather than real. The model inversion option is wrong because it requires evidence of an external threat actor reconstructing training data. The prompt injection option is wrong because it requires evidence of a malicious prompt injection from a competitor who gained access to the employee portal.