Skip to content
AI Security

Lesson 4 of 6 · 6 min read

Sensitive Information Disclosure

Sensitive information disclosure, LLM02 in the OWASP Top 10 for LLM Applications (2025 and 2026), covers personal data, secrets, proprietary information and other users' data leaking through an LLM application. The leak paths include retrieval without authorization, context shared across users, prompt and response logs, caches, fine-tuning data, model memorisation, third-party providers and the answer itself. Controls minimise what enters each path and enforce access before data reaches the model.

Reviewed By the AISecurityInterviews editorial team
free sign-in, no card required

Practise it as an interviewer asks it

Adversarial ML & Model PrivacyWhat is membership inference, and why does overfitting raise the risk?Mobile Application SecurityHow should a mobile app store tokens and keys on Android and iOS, and what does hardware-backed storage really protect?AI Red Teaming, Evaluation & ResponseHow would you detect and investigate an AI security incident?How LLM Systems WorkPrompting, RAG or fine-tuning on customer data: how does the security risk differ between them?LLM Application SecurityWhere can sensitive data leak in an LLM application?AI Red Teaming, Evaluation & ResponseHow do you respond to suspected cross-tenant data leakage through an LLM feature?

Course lessons that teach this

Securing an LLM Feature, End to EndPrioritize attack paths using reach, exposure and evidence20 min

Related lessons

RAG SecurityData Exfiltration Channels in LLM ApplicationsMembership Inference AttacksConfidentiality, Integrity, Availability and AuthenticitySystem Prompt Leakage