Shipped: The Applied AI Universe — Attacking 33 AI Models

Book 2 of The Applied AI Universe Coding Guide is an open-source, code-first educational resource demonstrating how adversarial attacks affect 33 AI model examples, spanning symbolic AI, classical machine learning, neural networks, deep learning, generative AI, and hybrid quantum computing.

The repository contains two executable notebooks covering 27 classical AI examples and six quantum/hybrid models. Each experiment reconstructs a model from Book 1, applies an attack suited to its architecture, and evaluates how model behavior changes under adversarial conditions.

Techniques include FGSM, PGD, black-box evasion, data poisoning, membership inference, model backdoors, reward poisoning, and quantum parameter manipulation. Experiments record clean versus adversarial performance, attack-success metrics, visualizations, and potential mitigation approaches.

Developed for students, educators, AI developers, and cybersecurity researchers, this resource helps learners understand how AI models fail and why adversarial testing matters before deployment.

All experiments use controlled educational models and public benchmark datasets. Quantum experiments are simulated on classical hardware.