Large language models (LLMs) are proving to be valuable tools for discovering zero-days, bypassing detection, and writing exploit code, thereby lowering the barrier to entry for penetration testers — ...