Large Language Models (LLMs) are susceptible to adversarial manipulations, despite employing advanced alignment strategies and content moderation mechanisms. Researchers have developed a genetic algorithm-based approach, GAS-Leak-LLM, to optimize suffixes for black-box LLM jailbreaking, highlighting the vulnerabilities of these models. This approach exploits the weaknesses in LLMs, allowing for the generation of harmful or policy-violating outputs. The GAS-Leak-LLM method demonstrates the potential risks associated with LLMs, which can have significant security implications1. As LLMs continue to evolve and become more pervasive, the risk surfaces associated with these models will also expand. The development of GAS-Leak-LLM serves as a reminder that the security of LLMs is a pressing concern that requires immediate attention. The ability to manipulate LLMs has significant consequences for the security and integrity of AI systems, so the development of effective countermeasures is crucial to mitigate these risks.