Model Risks in LLM
Lack of Transparency

Complex black-box models are vulnerable to undetected errors and attacks.