7 / 70 · 07 Security Testing for AI Apps · Insecure Output Handling and Model Denial of Service← prev⊞ allnext →☰ Read as one page
2.1LLM02: Insecure Output Handling
LLM output is often trusted and used in downstream operations without validation. When the model's output is interpolated into SQL queries, shell commands, HTML rendering, or API calls, it creates injection vectors that bypass traditional input validation (which only sanitizes user input, not AI output).