AI Security in the Era of Large Language Models: Red Teaming, Jailbreaking, and Prompt Injection | Synapse