Secure AI Project

Module 1 - Inside Artificial Intelligence: The Logic of Machines

Focus: Code-level introduction to AI logic​

Learning Goals:​

• Understand the structure of AI programs.​

• Learn about datasets, patterns, and training processes.​

• Write and test a simple rule-based AI program in Python.​

Activities:​

• Code a simple “chatbot” responding to greetings or moods.​

• Visualize data and make rule-based predictions.​

• Discussion: “What makes this intelligent?”​

Career Link: Software developer, data engineer, AI researcher.

Links: 

• Coding Example 1: Chatbot

• Coding Example 2: Gomoku

• Coding Example 3: Mind Reader

Module 2 - Machine Learning from Scratch

Focus: Programming simple machine learning models.​

Learning Goals:​

• Grasp supervised vs. unsupervised learning.​

• Use basic Python libraries (pandas, scikit-learn).​

• Train a simple classifier using open datasets (e.g., handwriting, emojis, or plant species).​

Activities:​

• Students load a dataset, split train/test data, and evaluate accuracy.​

• Visualize bias and discuss overfitting.​

• Reflection: “What happens when training data is unfair?”​

Career Link: Data scientist, ML engineer, quantitative analyst.​

Links:

• Coding Example 1: House Price Prediction

Module 3 - The Hidden World of Cybersecurity​

Focus: Coding for security and understanding encryption.​

Learning Goals:​

• Understand how encryption and hashing protect data.​

• Write Python scripts that encrypt and decrypt messages.​

• Explore brute-force attacks and password strength.​

Activities:​

• Needs to be planned​

• Debate: “Should third-party companies have access to encrypted data?”​

Career Link: Cybersecurity engineer, cryptographer, penetration tester.​

Module 4 - Networks, Firewalls, and Digital Defense

Focus: Simulating digital communication and protection mechanisms.​

Learning Goals:​

• Learn how networks exchange packets.​

• Understand how firewalls and intrusion detection work.​

• Use Python to simulate network messages and detect anomalies.​

Module 5 - AI Ethics, Adversarial Attacks, and Defense​

Focus: Coding the vulnerabilities of AI systems.​

Learning Goals:​

• Understand adversarial AI and how small changes can fool systems.​

• Implement a simple image-classification attack.​

• Explore how fairness and ethics connect to technical robustness.​

Activities:​

• Modify a trained ML model’s input image and see the misclassification.​

• Reflect: “When security meets bias, what could go wrong?”​

• Document solutions to mitigate vulnerabilities.​

Career Link: AI safety researcher, ethical hacker, AI policy specialist.​

Module 6 - Capstone Project: Build Your Own “Secure AI”​

Focus: Integrating AI and cybersecurity through programming.​

Learning Goals:​

• Design, code, and document a small AI-driven system with security awareness.​

• Present the logic and ethical choices behind it.​

• Reflect on career pathways in AI and cybersecurity.​

Activities:​

• Examples:​

• Spam detector with password-protected admin dashboard.​

• Image classifier that logs secure user access.​

• Chatbot that detects phishing messages.

Career Link: Full-stack developer, cybersecurity software engineer, AI researcher.​