Evaluate prompt injection against AI completion models
Engage in multimedia chat with LLMs and ML models