Your organization is deploying a customer-facing chatbot powered by a third-party LLM. The product team wants to connect it directly to the order management database to answer real-time inventory questions. As the security architect, what is the BEST design control?
A. Implement input validation to block prompt injection attempts
B. Place an API gateway with strict allow-listed queries between the LLM and the database
C. Require TLS 1.3 for all traffic between the chatbot and backend systems
D. Deploy a WAF tuned for LLM-specific attack signatures