Security Assertions By Large Language Models