The Emerging Attacks on Large Language Models (LLMs)

"key attack vectors that threat actors can exploit to compromise or manipulate LLMs".