AI Literacy Series

Course Information:
 

Course Title: AI Literacy Series

Year: 2025 - 2026

Target Audience: Form 4 or above

Prerequisite(s): Nil

Learning Hours:

• Workshop: 15.0

• Project: 4.0

​

Course Description:

This series provides an in-depth exploration of key artificial intelligence (AI) concepts and technologies. Through a mix of clear explanation and hands-on activities, each session builds a comprehensive understanding of a key area of artificial intelligence, fostering true AI literacy.
 

Course Learning Outcomes:

1. Explain how AI works: Demonstrate a working knowledge of fundamental AI concepts and how they operate.

2. Interact effectively with AI systems: Communicate and collaborate with AI tools to enhance your productivity and creativity.

3. Critically evaluate AI systems: Utilize AI technology responsibly with confidence, understanding its capabilities and limitations.

​

Topics:

• L1: AI For Every Mind

• L2: Transdisciplinary Nature of AI and Society

• L3: Introduction to Machine Learning and Large Language Model

• L4: AI in the Real World: Physical AI and Computer Vision

• L5: AI in Action through Intelligent Drones

• L6: Rise of Artificial Neural Network and Deep Learning

• L7: Generative AI for Images and Videos

• L8: The Art of the Prompt: Effective Communication with AI

• L9: AI in Healthcare: Diagnosis and Trust

• L10: AI in Laws and Humanities

• L11: Ethical Landscape and Implications of AI

• L12: Presentation and Exhibition of AI Projects

 

L1: AI For Every Mind

This session serves as an engaging introduction to the world of Artificial Intelligence, demystifying its core concepts and showcasing its real-world applications to build a foundational understanding for all learners. Students will explore:

• Examples of AI Applications: Introduce the rapid rise and growing capabilities of AI through the sharing of Large Language Models (LLMs), Visual Language Models (VLMs), Neural Signal-Operated Robots and autonomous vehicles and drones.

• The "Why": AI Literacy as an essential core competency for the future.

• Course Objectives: To enable learners to Explain, Interact with, and Critically Evaluate AI systems.

• Game Show: A Kahoot quiz to test what students know.

Goal: To spark curiosity and frame AI as an accessible and transformative tool that is relevant to "every mind."

 

L2: Transdisciplinary Nature of AI and Society

This session moves beyond the technology itself to explore the profound and complex relationship between AI and society. We examine the narratives surrounding AI, its foundational building blocks, and its vast implications across different fields, emphasizing that understanding AI is a transdisciplinary necessity. Students will explore:

• Narratives & Reality: Analyse the dystopian, utopian, and hype-driven narratives, grounding the discussion in the current reality of AI as a powerful tool that requires literacy.

• AI's Building Blocks: Introduced the core components that enable modern AI: Artificial Intelligence (the brain), Cloud Computing (the muscle), and Data (the fuel).

• A Transdisciplinary Tour: Explored concrete applications of AI across diverse sectors:

  • Healthcare: Medical diagnosis and sleep tracking.

  • History: Document analysis and biographical research.

  • Military & Government: Autonomous systems and AI-powered public services

  • Education: Automated grading and personalized learning tools.

• Societal & Ethical Challenges: Debate critical implications, including:

  • Privacy & Surveillance

  • Employment & Economic Inequality

  • Human Agency & Uniqueness

  • AI in Warfare: Peace vs. Conflict

• The Ultimate Responsibility: Conclude that AI is an amplifier of human intent, and its future depends on our wisdom, ethics, and governance.

Goal: To frame AI not just as a technical field, but as a societal force that demands critical evaluation and responsible stewardship from all disciplines.

 

L3: Introduction to Machine Learning and Large Language Model

This session provides a foundational understanding of the core technologies driving modern AI: Machine Learning (ML) and Large Language Models (LLMs). Through hands-on activities and clear analogies, we explore how machines learn from data and how they process human language. Students will explore:

• The AI Hierarchy: The relationship between Artificial Intelligence, Machine Learning, and Deep Learning.

• Core ML Concepts:

  • Supervised Learning: Training models with labelled data.

  • Classification vs. Regression: Predicting categories (e.g., "cat or dog") vs. continuous values (e.g., "exam score").

• Hands-On Practice: Build and test a simple image classification model using Teachable Machine to understand model training, confidence scores, and evaluation.

• Introduction to Large Language Models (LLMs):

  • LLMs as a type of deep learning model for language.

  • Explore examples like DeepSeek, ChatGPT, Llama,and Qwen.

  • Core Mechanism: The concept of next-word prediction.

  • Deterministic vs. Stochastic Output: Explain how the "Temperature" setting controls creativity vs. predictability in an LLM's responses.

• Interactive Activity: "Become a Human LLM" to experientially understand the impact of temperature on text generation.

Goal: To demystify how ML and LLMs work at a conceptual level, moving from abstract theory to practical, hands-on understanding and critical evaluation of AI outputs.

​

L4: AI in the Real World: Physical AI and Computer Vision

This session explores the world of Physical AI, where intelligent systems perceive and interact with the physical world. We delve into the sensors that enable perception, the AI models that understand visual data, and the methods these systems use to learn and make decisions. Students will explore:

• What is Physical AI? Intelligent that learn, adapt, and act in the real world.

• The Perception Stack:

  • Sensors: How AI "sees" the world using RGB Cameras, Stereo Cameras, LiDAR, Radar, Ultrasonic sensors, and GNSS/IMU.

  • Concept of Sensor Fusion: Combining data from multiple sensors to create a holistic digital representation of the environment.

• Computer Vision & CNNs:

  • Core Concept: How Convolutional Neural Networks (CNNs) work by detecting patterns—from simple edges to complex shapes—through a process called convolution.

• From Vision to Understanding:

  • Vision Language Models (VLM): AI that combines vision and language to perform commonsense reasoning about a scene (e.g., predicting what a cat might do next).

• Learning to Act:

  • Reinforcement Learning: How AI learns optimal behaviours through trial and error in a simulation, guided by a reward function.

Goal: To understand how AI perceive, reason, and act in our physical environment, bridging the gap between data and real-world interaction.

 

L5: AI in Action through Intelligent Drones

This session brings the concepts of Physical AI to life by exploring how drones integrate various AI technologies for autonomous flight, navigation, and interaction. We will move from theory to practice, examine specific AI models through hands-on activity. Students will explore:

• The "Eyes" of the Drone (Perception):

  • Bounding Box Detection (YOLO): Real-time bounding box object detection and tracking.

  • Occupancy Grid Mapping: LiDAR/vision-based environment modelling

• The "Gestures" of Control:

  • Pose Detection: Skeleton tracking for gesture recognition

• Hands-On Learning: Live demonstration and participant interaction showcasing the integrated AI features.

Goal: To demonstrate how AI models like object detection and spatial mapping are integrated into drones to enable real-world autonomous flight and human interaction.

 

L6: Rise of Artificial Neural Network and Deep Learning

This session dives into the architecture and mechanics of the technology that powers modern AI: Artificial Neural Networks (ANNs) and Deep Learning. Through relatable analogies and a hands-on exam score prediction example, students will explore how neural networks learn from data, adjust themselves, and make predictions—mirroring the way humans learn and improve. Students will explore:

• From Biological to Artificial:

  • Introduce the structure of an Artificial Neural Network: Input Layer, Hidden Layers, and Output Layer.

• How Neural Networks Learn:

  • Forward Propagation: How data flows through the network via synapses and activation functions.

  • Weights as “Knobs”: Understand weights as adjustable parameters that define the network’s behaviour.

  • Training with Data: Use a simple exam score prediction example to illustrate the training process.

  • Error and Cost Functions: Measure how “wrong” the network’s predictions are.

  • Gradient Descent & Backpropagation: Learn how networks efficiently adjust weights by minimizing error—without brute force.

  • Why GPUs Matter: Explore the role of specialized hardware (e.g., NVIDIA GPUs) in training deep networks.

• Hands-On Lab: Build and train a simple neural network using Python in Google Colab.

Goal: To demystify how neural networks function at a foundational level, empowering students to understand the “how” behind AI’s learning capabilities and inspiring confidence to engage with hands-on model building.​

​

L7: Generative AI for Images and Videos

This session explores the frontier of AI creativity: how machines can generate, enhance, and manipulate visual content. From understanding the nature of creativity to using state-of-art AI that produce stunning imagery, students will uncover the technology behind today’s most impressive visual AI—and confront its ethical implications. Students will explore:​

• What is Creativity?

  • Redefine creativity in the age of AI as the ability to connect distant concepts and generate novel, valuable outputs.

• How AI Learns from Visual Data:

  • Training Datasets: ImageNet, LAION-5B (image-caption pairs).

  • Challenge: Generating coherent images from text requires deep semantic understanding.

• Key Technologies of Visual Generation:

  • Generative Adversarial Networks (GANs): Two competing neural networks (Generator & Discriminator). Its applications include super-resolution, photo Inpainting, and style transfer.

  • Diffusion Models: Unpack the step-by-step process of iterative denoising from random noise to structured output.

  • ControlNet for Precision Control: Conditional generation using sketches, poses, or edges as input. Its applications include generate from sketch, pose and motion transfer, and edge detector.

• Challenges in Video Generation: Temporal consistency, physics simulation, fine-detail preservation.

• Ethical and Societal Implications: “To see no longer means to believe.”

  • Deepfakes: AI can create convincing fake videos and the pose risks to truth and trust.

  • Copyright & Creativity: Debate whether AI-generated content infringes on intellectual property and whether AI enhances or diminishes human creativity.

Goal: To equip students with a technical and ethical understanding of how AI generates and manipulates visual media, empowering them to critically evaluate AI-generated content and responsibly engage with creative AI tools.

​

L8: The Art of the Prompt: Effective Communication with AI

This session transforms students from passive users into skilled communicators with AI, focusing on the critical skill of prompt engineering. By understanding how Large Language Models (LLMs) work and mastering structured techniques to guide them, we will unlock more accurate, creative, and useful AI responses—while staying firmly in control of the interaction. Students will explore:​

• Understanding Your AI Partner:

  • Recap LLMs as “super auto-complete” systems built on pattern recognition and stochastic next-word prediction.

  • Recognize key limitations: hallucinations, knowledge cutoffs, and context windows.

• The Core Principle: Better Input = Better Output

• Learn the foundational framework of effective prompting:

  • Role

  • Task/Goal

  • Context

• Prompting Techniques for Precision & Creativity:

  • Multi-Shot Prompting

• Structured Reasoning Techniques:

  • Chain-of-Thought (CoT)

  • Tree of Thoughts (ToT)

• Enhancing Accuracy with External Knowledge:

  • Retrieval Augmented Generation (RAG)

• Critical Engagement & Ethical Reflection:

  • Loss of Creativity?: Debate whether AI enhances or replaces human originality in writing and thinking.

  • The Human-in-the-Loop Approach: Practice starting with your own ideas first, then using AI as a brainstorming partner to refine and expand—ensuring you remain the driver of creativity.

Goal: To equip students with the strategic communication skills needed to interact effectively, critically, and creatively with AI systems, empowering them to leverage AI as a collaborative tool while maintaining originality.

​

L9: AI in Healthcare: Diagnosis and Trust

This session explores the transformative role of Artificial Intelligence in healthcare, moving from consumer wellness to life-critical clinical decision support. We will examine specific applications, crucially address why healthcare demands exceptionally rigorous AI evaluation compared to other fields, and learn how to evaluate their performance with a critical eye. Students will explore:

• Preventive & Personal Health: AI in Wearable Devices (e.g., ECG, Sleep Analysis).

• AI-Assisted Diagnosis: The role of Large Language Models (LLMs) in synthesizing medical information.

• Critical Care & Monitoring: Using LSTM models for real-time patient surveillance in the ICU.

• The Crucial Step: Evaluation

  • Why Healthcare Demands Higher Standards

  • Binary Classification Metrics (TP/TN/FP/FN, Accuracy, Precision, Recall, F1-Score, Specificity)

Goal: To explore the critical applications of AI in healthcare and master the fundamental statistical framework for evaluating binary classification models to ensure their safety and efficacy.

​

L10: AI in Laws and Humanities

This session explores the transformative impact of Artificial Intelligence beyond the sciences, delving into its powerful applications in the legal system and humanities. Students will discover how AI serves as a bridge between data and human understanding, unlocking historical insights, enhancing legal analysis, and revealing cultural patterns—while engaging in critical discussions about ethics, bias, and the enduring role of human judgment. Students will explore:

• Legal AI Systems: How neural networks are trained on case law databases to predict outcomes and identify precedents (e.g., HKU AI Lawyer for drug trafficking sentencing).

• AI in Text & Document Analysis: Use of Computer Vision and NLP to digitize and analyse historical documents (e.g., Qing Dynasty exam papers) where traditional OCR fails.

• Cultural & Media Pattern Recognition: Application of colour analysis and thematic extraction from visual and textual media to uncover cultural trends.

• Ethical Implications and Human Responsibility:

  • Bias & Fairness: Could an AI trained on historical judgments perpetuate past biases?

  • Job Transformation: Will AI replace lawyers and historians, or augment their expertise?

  • Trust & Authenticity: How do we verify AI’s predictions or historical analyses?

• Reinforce the core principle: AI is a tool, not a substitute for human critical thinking, empathy, and ethical judgment.

Goal: To demonstrate how AI serves as a powerful interdisciplinary bridge—connecting law and data, past and present, art and science—while empowering students to think critically about its ethical use and the irreplaceable value of human insight.

​

L11: Ethical Landscape and Implications of AI

This session explores the ethical questions and societal responsibilities raised by Artificial Intelligence. Students will examine key philosophical debates, real-world dilemmas, and emerging governance challenges to cultivate a critical and human-centred perspective on AI development and deployment. Students will explore:

• Philosophical Foundations: Discuss the Turing Test, Weak vs. Strong AI, Narrow AI vs Artificial General Intelligence (AGI) and Searle's Chinese Room Argument to probe the nature of intelligence, understanding, and consciousness.

• Ethical dilemmas of AI: Evaluate the ethical trade-offs of AI in education, hiring, autonomous vehicles and weapon systems.

• Ethical Frameworks in Practice: Apply normative ethics (Utilitarianism vs. Deontology) to AI dilemmas, featuring the Moral Machine experiment.

• Hidden costs of AI: Examine the environmental footprint, risks of inequality and centralisation of power.

• Regulatory challenges: Analyse issues of transparency and accountability.

• Global Governance: Consider the need for international coordination to ensure safe, fair, and equitable AI development.

Goal: To equip students with the critical frameworks needed to navigate the ethical complexities of AI, empowering them to advocate for responsible innovation and equitable outcomes in an AI-driven world.

​

L12: Presentation and Exhibition of AI Projects

This final session serves as the capstone event of the AI Literacy Series, where participants present and exhibit their group projects. Teams will showcase their AI topic posters, demonstrating comprehensive understanding and application of course concepts through structured presentations and engaged discussion. This session emphasises knowledge synthesis, critical reflection, and collaborative learning, providing a platform for participants to articulate insights, receive feedback, and celebrate their journey in AI literacy.