Towards Real-World Adversarial Examples In Ai-Driven Cybersecurity