Security researchers demonstrate that large language models can effectively augment the discovery and exploitation of memory corruption bugs within the FreeBSD kernel. While these tools successfully streamline vulnerability research, expert human intervention remains necessary to navigate complex isolation mechanisms like jails.