Curriculum alignment · The course
Grade 7: units, topics and lessons
Every Grade 7 topic in the LittleAIMaster AI course, its lessons, and the requirements from 14 AI curricula it helps meet. Open a topic to see its lessons; lessons cited as evidence are in bold.
6units61topics1,680lessons233requirements met
Looking for the learner view? See the Grade 7 learning path.
Unit 1: Patterns Everywhere 10 topics
Welcome AI Investigators18 lessons· meets 2 requirements inCSTA (US)AI4K12 (US)
Lessons
- Maya's Notebook Question
- Your Journey So Far
- The Secret Investigators Chase
- The Real Map for Your Year
- Level Up Match
- Four Different Inventions, One Job
- Your Investigation Kit
- Order the Steps
- Investigators Notice Before They Explain
- Nobody Writes the Rules
- Sort the Thinking
- Investigator Power Meter
- Your Photo App Runs on Patterns
- Investigator Radar
- Checkpoint Quiz
- What Makes an AI Investigator
- Investigator Words
- AI Investigator!
Requirements this topic helps meet
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
Patterns Rule the World18 lessons· meets 2 requirements inCSTA (US)AI4K12 (US)
Lessons
- The Pattern Hunt
- Three Kinds, One Test
- The Hidden Rule Behind So Many Things
- True or False: Nature
- Nature Match
- Patterns Appear Across Totally Different Worlds
- Break the Rule
- Sequence Master
- Prediction Starts With Good Pattern Memory
- Patterns Everywhere
- Pattern Sorter
- Pattern Strength Lab
- Spotify Learns From Listening Patterns
- Pattern Detective
- Pattern Quiz
- Patterns Really Do Rule the World
- Pattern Words
- Pattern Master!
Requirements this topic helps meet
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
Your Pattern-Finding Brain18 lessons· meets 5 requirements inUAECSTA (US)AI4K12 (US)
Lessons
- Raj's Secret Power
- How a Pattern Gets Built
- The Answer With No Working Out
- Two Brain Experiments
- Brain Skill Match
- Learning From Almost Nothing
- Brain vs AI: Three Rounds
- Sort the Thinking
- The Easy Things Are the Hard Things
- Learned, or Built In?
- Brain Growth Order
- Brain Learning Meter
- Face Unlock Copies One Slice of You
- Pattern Speed Test
- Brain Quiz
- Pick a Round Your Brain Won
- Brain Vocabulary
- Brain Champion!
Requirements this topic helps meet
- UAECompare AI with humans (Cycle 2 part)
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from identifying the five human senses and sensory organs (K-2), to comparing how different animals perceive the world differently than humans, e.g. sonar, ultraviolet vision, colorblindness (3-5), to giving examples of how humans combine information from multiple sensory modalities to resolve ambiguous signals (6-8); AI4K12 marks 9-12 N/A for this topic since it is considered adequately covered by the lower bands (leaving deeper biology/psychology to other courses).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from giving commonsense explanations of human behavior useful for a robot to know, e.g. choosing a plausible story ending (K-2), to explaining what everyday cultural and naive-physics knowledge a computer needs to understand a story (3-5), to explaining the naive-physics inference rules (mass, volume, conservation, forces) needed to solve a physics reasoning problem (6-8), to explaining the cultural, naive-physics, and folk-psychology knowledge needed to interpret a fable, fairytale, or Winograd-schema sentence (9-12).
Classification Basics18 lessons· meets 3 requirements inUAECSTA (US)AI4K12 (US)
Lessons
- The Floor Maya Could Not See
- How Grouping Actually Works
- Two Rooms, Two Sets of Piles, One Finished
- Groupings You Already Live In
- What Grouping Buys You
- One Good Group Means You Never Start From Zero
- Three Shapes of Classification
- The Four Moves, In Order
- The Tomato Argument Nobody Wins
- Where AI Is Doing This Right Now
- Name That Shape
- Grouping Dials
- Your Inbox Was Sorted Before You Woke Up
- Judge the Grouping
- Classification Checkpoint
- Make the Case
- The Words For It
- Classification Pro!
Requirements this topic helps meet
- UAEUse generative AI models to create new content
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
Features and Attributes20 lessons· meets 13 requirements inADEKEngland + CambridgeCSTA (US)AI4K12 (US)+2
Lessons
- The Mystery Fruit
- Feature or Category?
- One Clue, Three Answers
- Five Kinds of Features
- Feature Match
- A Few Smart Features Can Beat Hundreds of Weak Ones
- Strong, Weak and Misleading
- Strong Clue or Weak Clue?
- Your Brain Throws Most of It Away
- What AI Actually Measures
- AI Feature Steps
- Features Become a Map
- Find the Nearest Neighbour
- Feature Finder Lab
- Face Unlock Measures Key Details
- Feature Detective
- Features Quiz
- Which Features Matter Most
- Feature Words
- Chapter 5 Complete
Requirements this topic helps meet
- ADEKBuild a chart/dashboard
- England + CambridgePlan statistical investigations, design databases/spreadsheets with cell referencing and functions, and know that data is used to solve problems across industries
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- CSTA (US)Create and evaluate different abstract representations (e.g., subway map).
- CSTA (US)Create and evaluate an appropriate AI algorithm (e.g., a decision tree classifier) to accomplish a task.
- AI4K12 (US)Progress from identifying the five human senses and sensory organs (K-2), to comparing how different animals perceive the world differently than humans, e.g. sonar, ultraviolet vision, colorblindness (3-5), to giving examples of how humans combine information from multiple sensory modalities to resolve ambiguous signals (6-8); AI4K12 marks 9-12 N/A for this topic since it is considered adequately covered by the lower bands (leaving deeper biology/psychology to other courses).
- AI4K12 (US)Progress from giving examples of visual features (parts, textures, colors) used to recognize an object class such as cats (K-2), to illustrating how face detection extracts facial features (3-5), to simulating an edge detector to illustrate feature extraction from images (6-8), to explaining how features (formants, edges, intensity gradients) are extracted from waveforms and images at multiple scales (9-12).
- AI4K12 (US)Progress from constructing a map of a home, school, or neighborhood and understanding that 'the map is not the territory' (K-2), to demonstrating how different map styles (geometric vs. topological) capture different aspects of the world (3-5), to showing how a game board (tic-tac-toe, chess) can be represented as a plain-language description of squares, relationships, and piece positions (6-8), to describing how to represent a concept as a schema.org-style schema with superconcepts and inherited properties (9-12).
- AI4K12 (US)Progress from identifying the features that make objects in a collection unique and organizing them into a feature table (K-2), to constructing a feature-vector representation for a set of objects and showing that similar objects lie close together in feature space (3-5), to explaining how word embeddings represent words as numeric vectors for tasks like machine translation (6-8), to describing how a transformer network maps sequences of word-embedding vectors to output sequences (9-12).
- AI4K12 (US)Progress from analyzing a decision-tree-building game and how the learning algorithm adds/replaces nodes (3-5), to comparing how decision-tree learning (add a node) differs from neural-network learning (adjust every weight) (6-8), to describing how decision-tree, neural-net, regression, and reinforcement-learning algorithms each adjust their own type of internal representation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- UNESCOAcquire age-appropriate technical skills in AI programming, understanding how supervised, unsupervised and reinforcement-learning algorithms scrape/process data, train, and function (CG4.2.3.2).
- Australia + NZUnderstand how mathematical skills and concepts — pattern recognition, spatial reasoning, chance, data and algorithms — are applied to train AI systems
Simple vs Complex Patterns18 lessons· meets 2 requirements inCSTA (US)AI4K12 (US)
Lessons
- The Puzzle Race
- Pattern Secrets
- The Puzzle Ms. Sharma Promised to Fix
- Signs of a Simple Pattern
- Pattern Match
- A Small Increase in Complexity Can Change Everything
- Complexity Levels
- Order by Complexity
- Hard Patterns Are Not Always Big Patterns
- AI & Complexity
- Sort by Complexity
- Complexity Meter
- Weather Apps Tackle Complex Patterns
- Complexity Detective
- Complexity Quiz
- Why Complex Patterns Are Hard
- Complexity Words
- Complexity Master!
Requirements this topic helps meet
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
Prediction from Patterns21 lessons· meets 5 requirements inUAEADEKCSTA (US)AI4K12 (US)+1
Lessons
- The Weather Prediction Challenge
- 5 Steps of Prediction
- The Clue That Turns a Guess Into a Prediction
- Guess or Prediction?
- Match Patterns to Predictions
- Tiny Clues Can Predict Big Outcomes
- Prediction Confidence Levels
- AI Prediction in Action
- Good Predictions Depend on Repeated Proof
- What AI Learned, What AI Predicts
- Strong vs Weak Predictions
- Predict, Then Pick
- Safe or Wild?
- Three Runs, One Favourite
- Prediction Confidence Meter
- Google Maps Predicts Trips From Past Traffic
- When Predictions Go Wrong
- Prediction Power Quiz
- What Makes a Prediction Strong
- Prediction Vocabulary
- Prediction Expert!
Requirements this topic helps meet
- UAEUse generative AI models to create new content
- ADEKHuman/AI-generated art + ethics
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- Australia + NZUnderstand how mathematical skills and concepts — pattern recognition, spatial reasoning, chance, data and algorithms — are applied to train AI systems
Patterns in Data18 lessons· meets 8 requirements inUAEADEKCBSECSTA (US)+3
Lessons
- The Class Test Mystery
- 5 Types of Data Patterns
- The Secret Hidden in Plain Numbers
- Three Shapes of Data
- Match Data Sources to Patterns
- Data Can Reveal Things No One Saw Directly
- The 5 Steps AI Takes
- Which Step Is the AI Doing?
- A Pattern Needs a Good Question Too
- Data Patterns in Your Life
- Good vs Misleading Patterns
- Data Insight Meter
- Spotify Finds Patterns in Massive Listening Data
- Data Pattern Detective
- Data Patterns Quiz
- How Data Becomes Insight
- Data Vocabulary
- Data Detective!
Requirements this topic helps meet
- UAELearn from data; algorithms make decisions
- ADEKBuild a chart/dashboard
- CBSEDistinguish key predictive techniques - regression, classification, clustering - and explain key AI domains: Data Science, Computer Vision, Natural Language Processing; explain uses of AI in healthcare, education, transport, communication (Class 7).
- CSTA (US)Collect data using computational tools and transform the data to make it more useful and reliable.
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- UNESCOAcquire age-appropriate knowledge and skills on datasets — applying tools/languages to acquire, clean and transform data into a form suitable for storing, processing and analysing (e.g. SQL, NoSQL, SparkSQL, Apache Flink) (CG4.2.3.1).
- SingaporeDevelop data competencies as one of the nine core NDLP digital competencies.
Unit Review20 lessons· meets 6 requirements inCSTA (US)AI4K12 (US)Korea, Japan, HKAustralia + NZ
Lessons
- Your Pattern Journey
- Your Learning Journey
- Rules or Examples?
- The One Idea Hiding Inside the Whole Unit
- Concept Connections
- So Many AI Tools Depend on the Same Core Skill
- Vocabulary Power
- AI Applications
- Reviewing Helps Patterns Stick
- Cause, or Just Linked?
- 5 Big Ideas
- What Makes a Pattern Easy or Hard
- YouTube Uses Every Unit 1 Skill at Once
- Human vs AI
- Raise Raj's Waste Sorter
- True or False?
- Unit 1 Final Quiz
- What Unit 1 Really Taught Me
- Unit 1 Vocabulary
- Unit 1 Champion!
Requirements this topic helps meet
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- CSTA (US)Identify the kinds of AI models (e.g., classifier, predictor, recommender) people interact with in their daily lives.
- AI4K12 (US)Progress from identifying a problem as a classification problem or a search problem (K-2), to categorizing problems as classification vs. search using examples like Teachable Machine (3-5), to categorizing problems as classification, prediction, combinatorial search, or sequential decision problems (6-8), to categorizing real-world problems across eight reasoning-problem types including heuristic search, adversarial search, logical deduction, and statistical inference (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- Korea, Japan, HKDistinguish skill-based reasoning (direct task response), rule-based reasoning (decisions from learned patterns) and knowledge-based reasoning (applying machine learning to find the best solution) in a hands-on robot-reasoning experiment (e.g. a moral-dilemma or line-following task).
- Australia + NZUnderstand how mathematical skills and concepts — pattern recognition, spatial reasoning, chance, data and algorithms — are applied to train AI systems
Pattern Challenge24 lessons· meets 8 requirements inCSTA (US)AI4K12 (US)UNESCOSingapore+1
Lessons
- The Case on the Board
- Six Skills, One Case
- The Investigator's Checklist
- The Clue Everyone Walked Past
- Challenge 1: Name the Pattern
- One Toolkit, Jobs That Look Nothing Alike
- Where Investigators Get It Wrong
- Sort the Patterns
- Many Small Clues Beat One Loud Clue
- Challenge 2: Good Classification?
- The Canteen Case
- Match the Case to the Skill
- Maya's Five-Day Tally
- Challenge Readiness Meter
- Maps Runs Your Whole Toolkit at Once
- Order the Data Steps
- Challenge 3: Are These Good Features?
- Design an AI on Paper
- Challenge 4: Strong or Weak Prediction?
- Your Journey Through Unit 1
- The Questions With No Easy Answer
- Final Challenge Quiz
- My Strongest AI Skill
- Unit 1 Complete!
Requirements this topic helps meet
- CSTA (US)Collect data using computational tools and transform the data to make it more useful and reliable.
- CSTA (US)Use data analysis tools and techniques to identify patterns in data representing complex systems.
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- UNESCOAcquire age-appropriate knowledge and skills on datasets — applying tools/languages to acquire, clean and transform data into a form suitable for storing, processing and analysing (e.g. SQL, NoSQL, SparkSQL, Apache Flink) (CG4.2.3.1).
- SingaporeDevelop data competencies as one of the nine core NDLP digital competencies.
- Australia + NZUnderstand how mathematical skills and concepts — pattern recognition, spatial reasoning, chance, data and algorithms — are applied to train AI systems
Unit 2: Teaching Machines 10 topics
What is Training?18 lessons· meets 12 requirements inUAEADEKIBCSTA (US)+2
Lessons
- Teaching Robo to Sort
- What is AI Training?
- Humans vs AI Learning
- What Turns a Wrong Guess Into Learning?
- Learning Connections
- Nobody Ever Wrote Down the Rule for 'Camel'
- Programming vs Training
- Programming or Training?
- Forty of the Same Dog Teach Less Than Forty Different Ones
- Why Training Matters
- Training Process
- The Training Strength Meter
- The Spam Filter That Never Finishes Training
- Training or Not?
- Training Quiz
- Explain Robo to Someone Who Missed It
- Training Vocabulary
- Training Expert!
Requirements this topic helps meet
- UAELearn from data; algorithms make decisions
- ADEKTrained-on-examples / rule-vs-learned
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)AI content is integrated (not a separate strand): students learn rule-based vs. data-driven approaches, machine learning basics, and related societal impacts such as bias, within the Algorithms & Design strand.
- AI4K12 (US)Progress from describing and giving examples of how people vs. computers learn (K-2), to differentiating human flexible learning from computer pattern-finding/trial-and-error learning (3-5), to contrasting human learning characteristics with narrowly-defined machine learning problems (6-8), to defining supervised, unsupervised, and reinforcement learning and giving human-learning analogues for each (9-12).
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from demonstrating how training data are labeled by class in a tool like Teachable Machine (3-5), to explaining the difference between training a reasoning model and using it on new data (6-8), to illustrating all steps of building a classifier/predictor with machine learning, including choosing features, labeling, cross-validation, and test-set evaluation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from demonstrating how weights are assigned in a neural network to produce a desired input/output behavior (3-5), to demonstrating a learning rule (e.g. perceptron rule) that adjusts weights of a one-layer network in response to errors (6-8), to training a multilayer neural network with the backpropagation algorithm and describing how weights and hidden-unit outputs change during learning (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- OECD–EC (Europe)Identify the benefits and drawbacks of each AI approach for a task (Manage 2, intermediate).
Learning from Examples18 lessons· meets 6 requirements inADEKIBAI4K12 (US)
Lessons
- Raj's Pet Detector
- How AI Learns
- What AI Learns
- Why Wasn't One Photo Enough?
- Pattern Pairs
- AI Can Learn the Idea of Cat-ness
- Memorizing vs Learning
- Good vs Limited Examples
- Diverse Examples Teach More Than Perfect Copies
- How Many Examples Is Enough?
- Example Quality Order
- Example Quality Meter
- Your Camera Finds Faces From Labeled Examples
- Good Training Examples?
- Examples Quiz
- Learning Patterns, Not Just Pictures
- Example Vocabulary
- Example Expert!
Requirements this topic helps meet
- ADEKTrained-on-examples / rule-vs-learned
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
Good Examples vs Bad Examples18 lessons· meets 12 requirements inADEKIBCSTA (US)AI4K12 (US)+2
Lessons
- The 500 Photos That Taught Nothing
- What Makes a GOOD Example
- Camera Fix or Keyboard Fix?
- Why Can 500 Photos Still Fail?
- Cause and Effect Match
- Fewer Photos Can Teach More
- What Bad Data Does to an AI
- Keep It or Bin It?
- The Shortcut Trap
- What Actually Fixed It?
- Fix Raj's Dataset
- Data Quality Meter
- Why a Voice Assistant Understands Some People Better
- Spot the Bad Example
- Quality Quiz
- Explain It to Raj
- Quality Vocabulary
- Data Quality Master!
Requirements this topic helps meet
- ADEKExamine dataset gaps/bias
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- CSTA (US)Students build data fluency and examine the societal impacts of large-scale data collection used to train AI systems.
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- OECD–EC (Europe)Explain how biased data or design choices create unfair outcomes (Engage 6, intermediate).
- OECD–EC (Europe)Compare the importance of data selection and collection methods for an AI system (Shape 3, intermediate).
- Australia + NZConsider the possible bias or harm to others that may occur when inputting data into AI systems, and explore how AI systems use predictive strategies based on large amounts of data to perform tasks or solve problems
The Learning Loop25 lessons· meets 11 requirements inADEKIBAI4K12 (US)Korea, Japan, HK
Lessons
- Priya's Bird Quest
- The Loop Revealed
- Before or After the Guess?
- What Makes the Loop Actually Improve?
- Step Matcher
- One Small Correction Can Change the Next Guess
- Step 1: INPUT
- PROCESS: Which Clue Separates?
- One Clue, One Vote
- Feedback Turns Repetition Into Progress
- OUTPUT: How Sure Should It Be?
- Step 4: FEEDBACK
- Tune the Learning Loop
- A Music App Learns Your Taste One Loop at a Time
- Loop Sequencer
- Train Byte on Priya's Birds
- Gmail's Spam Filter
- Feedback Sorter
- Loop Superpowers — and the Catch
- Problem Solver
- Loop Detective
- Loop Quiz
- The Learning Loop in My Own Words
- Loop Vocabulary
- Loop Master!
Requirements this topic helps meet
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- AI4K12 (US)Progress from describing and giving examples of how people vs. computers learn (K-2), to differentiating human flexible learning from computer pattern-finding/trial-and-error learning (3-5), to contrasting human learning characteristics with narrowly-defined machine learning problems (6-8), to defining supervised, unsupervised, and reinforcement learning and giving human-learning analogues for each (9-12).
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from analyzing a decision-tree-building game and how the learning algorithm adds/replaces nodes (3-5), to comparing how decision-tree learning (add a node) differs from neural-network learning (adjust every weight) (6-8), to describing how decision-tree, neural-net, regression, and reinforcement-learning algorithms each adjust their own type of internal representation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from illustrating that a 1-3 neuron network is a function that computes weighted-sum/threshold outputs (3-5), to illustrating the layered structure (input/hidden/output) of a neural network and its transfer functions (6-8), to describing feed-forward, 2D convolutional, recurrent, and generative adversarial network architectures and their typical uses, e.g. deepfakes (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from demonstrating how weights are assigned in a neural network to produce a desired input/output behavior (3-5), to demonstrating a learning rule (e.g. perceptron rule) that adjusts weights of a one-layer network in response to errors (6-8), to training a multilayer neural network with the backpropagation algorithm and describing how weights and hidden-unit outputs change during learning (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- Korea, Japan, HKExplain how a neural network is built from perceptron units (inputs, weightings, bias, activation function) and how it "learns" via backpropagation to adjust weights to fit training data.
- Korea, Japan, HKUse a simulation tool (e.g. a traffic-light decision simulator) to explore how an AI system makes sequential decisions under simulated conditions.
Supervised Learning Basics33 lessons· meets 17 requirements inUAEADEKCBSEIB+3
Lessons
- Nani's Spotted Leaves
- What is Supervised Learning?
- Input or Label?
- The Two Piles Nobody Can Name
- Training Phase
- Fair Exam or Not?
- Every Label Is a Human Decision
- Concept Matcher
- One Wrong Label Bends Every Prediction After It
- Classification Explained
- Binary or Multi-class?
- Regression Explained
- Where the Numbers Come From
- Type Sorter
- What Actually Makes a Model Strong
- PayPal Learns Fraud From Payments People Already Judged
- Train Maya's Leaf Sorter Yourself
- Needs a Label, or Not?
- Learning Without Labels
- Group the Unlabelled Drawings
- A Third Way: Learning From Rewards
- Three Ways to Learn
- Real Classification Apps
- Real Regression Apps
- Pipeline Builder
- Why Teams Reach For It First
- The Hard Parts
- Pros & Cons Matcher
- Type Detective
- Supervised Learning Quiz
- Ten Labels Nani Got Wrong
- Key Vocabulary
- Supervised Learning Expert!
Requirements this topic helps meet
- UAELearn from data; algorithms make decisions
- ADEKExamine dataset bias, propose fixes
- CBSEDescribe the stages of the AI project cycle (Define Problem, Collect Data, Test AI Tools, Reflect and Improve); apply no-code tools to tackle real-world problems and reflect on their utility/effectiveness; analyse contributions of AI to fields like healthcare, automation, and education, understanding both benefits and risks; describe AI ethics as values/guidelines for responsible AI (Class 8).
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- AI4K12 (US)Progress from identifying a problem as a classification problem or a search problem (K-2), to categorizing problems as classification vs. search using examples like Teachable Machine (3-5), to categorizing problems as classification, prediction, combinatorial search, or sequential decision problems (6-8), to categorizing real-world problems across eight reasoning-problem types including heuristic search, adversarial search, logical deduction, and statistical inference (9-12).
- AI4K12 (US)Progress from describing and giving examples of how people vs. computers learn (K-2), to differentiating human flexible learning from computer pattern-finding/trial-and-error learning (3-5), to contrasting human learning characteristics with narrowly-defined machine learning problems (6-8), to defining supervised, unsupervised, and reinforcement learning and giving human-learning analogues for each (9-12).
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from demonstrating how training data are labeled by class in a tool like Teachable Machine (3-5), to explaining the difference between training a reasoning model and using it on new data (6-8), to illustrating all steps of building a classifier/predictor with machine learning, including choosing features, labeling, cross-validation, and test-set evaluation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from explaining how reinforcement learning lets a computer learn from trial-and-error experience with a reward signal, e.g. a grid-world navigation demo (3-5), to explaining the difference between supervised learning (told the correct output) and reinforcement learning (told only how well it's doing) (6-8), to selecting the appropriate learning-algorithm family (supervised, unsupervised, or reinforcement) for a given reasoning problem (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from illustrating why training a classifier for a broad concept like 'dog' requires large, diverse image data by browsing datasets such as ImageNet or Coco (3-5), to illustrating how objects in images are segmented and labeled with bounding boxes to build an object-recognition training set (6-8), to evaluating a real AI training dataset's size, acquisition/labeling method, storage cost, and time-to-produce, including where bias can enter (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- UNESCODevelop an age-appropriate understanding of the three types of AI algorithms — supervised learning, unsupervised learning, and reinforcement learning — and how the data behind them is acquired and labelled (CG4.1.3.2).
- UNESCOPractise skills to identify and scope a problem that could be solved by building or training a new AI model, based on a simulation project (CG4.1.4.2).
- UNESCODevelop skills to assess an AI system's needs for data, algorithms, programming languages, software, and computing hardware, including feasibility given regulatory/ethical restrictions and cost (CG4.1.4.3).
- UNESCOAcquire age-appropriate technical skills in AI programming, understanding how supervised, unsupervised and reinforcement-learning algorithms scrape/process data, train, and function (CG4.2.3.2).
- ChinaStudents understand the basic workflow of machine learning and the concept of supervised learning
Teaching by Showing35 lessons· meets 28 requirements inUAEADEKCISCE + statesIB+6
Lessons
- Emma's AI Project
- Your Job or Byte's?
- Why Did the Demo Fail?
- Teachable Machine
- A Browser Can Become a Custom AI Lab
- Which Project Type?
- Process Matcher
- A Copy of a Photo Teaches Nothing New
- Class List Check
- Step 2: Examples
- Example Sorter
- Build a Stronger Demo
- Teachable Machine Makes Classroom AI Possible
- Train and Test
- Step Sequencer
- The 5 Vs
- Train Emma's Rock vs Paper Yourself
- Mistake Matcher
- Project Ideas
- Other ML Tools
- Coin Classifier Challenge
- Fill Priya's Labels
- Training Detective
- Bias Hunt: Your Brief
- Bias Hunt: What to Look For
- Your Own Bias Investigation
- Bias Hunt Check
- Capstone: Your AI's Model Card
- Raise It, Then Read the Card
- Read the Model Card
- Defend Your Model Card
- Teaching Quiz
- Explain Teaching by Showing
- Teaching Vocab
- Training Expert!
Requirements this topic helps meet
- UAELearn from data; algorithms make decisions
- UAEEthics, cultural values and inclusion guide AI decisions
- UAEUse generative AI models to create new content
- UAEDesign AI projects for real-world problems
- ADEKTrain/test a model, report performance
- ADEKExamine dataset bias, propose fixes
- CISCE + statesDevelop computational-thinking and coding skills using block-based tools such as Scratch/Blockly through hands-on, project-based learning applied to real-world problems
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)Create and evaluate an appropriate AI algorithm (e.g., a decision tree classifier) to accomplish a task.
- CSTA (US)Create a program using available AI tools, AI plugins, APIs, and/or AI models, with ethical considerations for the model's end users and others impacted: fairness, bias, and accuracy, and then create a model card.
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from demonstrating how training data are labeled by class in a tool like Teachable Machine (3-5), to explaining the difference between training a reasoning model and using it on new data (6-8), to illustrating all steps of building a classifier/predictor with machine learning, including choosing features, labeling, cross-validation, and test-set evaluation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- AI4K12 (US)Progress from creating a model card for a classification or prediction model documenting its intended use, training data, accuracy, and limitations (3-5), to constructing an ethical matrix of stakeholders and their values for an AI system and evaluating alignment (6-8), to designing an AI system using a full ethical design process that considers all stakeholders' values across fairness, transparency, explainability, accountability, privacy, and societal values (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Identify unintended impacts of an AI use that differ from its immediate goals (Engage 7, intermediate).
- OECD–EC (Europe)Experiment with multiple AI tools to develop diverse ideas (Create 2, intermediate).
- OECD–EC (Europe)Be aware that AI can automatically create digital content (text, news, music, images) from existing content as its source, which can be difficult to distinguish from human creations, including in news media/journalism (AI examples 5 and 6, competence 3.1).
- OECD–EC (Europe)Be interested in experimenting with various types of AI systems depending on personal needs (e.g. virtual assistants, image analysis, speech/face recognition, robots) (AI example 71, competence 5.2).
- UNESCODiscuss age-appropriate real-world cases around six core AI ethical principles — do no harm, proportionality, non-discrimination, sustainability, human determination, and transparency/explainability — and their personal implications (CG4.1.2.2).
- UNESCOInternalise ethical principles on AI (human rights, data privacy, human agency, equity, inclusion, social justice, environmental sustainability) into reflective practices, e.g. advocating for inclusive AI interfaces and reporting discriminatory bias found in AI tools (CG4.1.2.3).
- UNESCOCategorise general safety risks of AI and typical AI incidents, deepen knowledge of human rights to data protection and privacy and creators' legal responsibility to collect data only with informed consent, and practise strategies to ensure personal data is collected, used, shared, archived and deleted only with consent (CG4.2.2.3).
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- SingaporeUse basic coding concepts (variables, loops, functions) in a block-based coding platform to build simple programs.
- Korea, Japan, HKExamine problems that can arise in the collection and use of data needed for AI learning (bias, privacy, consent) and propose legally, socially and ethically valid solutions ([9정04-05]).
- Korea, Japan, HKIdentify AI ethical principles and analyse ethical issues raised by real AI applications (bias, fairness, accountability), as a dedicated unit distinct from general 'societal impact'.
When AI Gets Confused24 lessons· meets 6 requirements inADEKIBAI4K12 (US)UNESCO
Lessons
- Raj's Study Fail
- Too Little, Too Much, Just Right
- Why Did Perfect Training Still Fail?
- Memorized or Learned?
- Smart AI Lives in the Middle
- The Red Sofa Cat
- Overfitting Matcher
- More Training Time Can Make Predictions Worse
- Overfitting Causes
- Fix It or Feed It?
- Find the Learning Sweet Spot
- Face Unlock Must Work Beyond One Room
- What is Underfitting?
- Underfitting Fixes
- Learning Spectrum
- Build It to Generalize
- Real-World Cases
- Stop or Keep Training?
- Problem Sorter
- AI Diagnostician
- Confusion Quiz
- Explain Overfitting
- Confusion Vocab
- Confusion Master!
Requirements this topic helps meet
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- UNESCOGain interdisciplinary foundational knowledge for AI (e.g. artificial neural networks, strong vs weak AI, connections to STEM, languages and social studies) and reflect on AI's reciprocal impact on those subjects (CG4.1.3.3).
Testing What AI Learned25 lessons· meets 9 requirements inUAEADEKIBCSTA (US)+3
Lessons
- A Score That Proves Nothing
- The Three Data Sets
- How Can 95% Still Be Dangerous?
- Train It, Then Test It
- Why Hide Some of the Data?
- Validation Set: The Weekly Quiz
- Seal It or Spoil It
- Match the Data Set Purpose
- High Accuracy Can Still Miss What Matters
- Work Out the Accuracy
- Tune the Testing Plan
- Gmail Depends on Careful Testing
- When Accuracy Lies
- When Can We Trust Accuracy?
- The Confusion Matrix
- Sort the Four Boxes
- Precision and Recall
- Precision vs Recall Examples
- The Trade-off Challenge
- Best Practices for Splitting Data
- Design the Data Split
- Testing Knowledge Quiz
- Explain the Test Set
- Testing Vocabulary
- Testing Expert!
Requirements this topic helps meet
- UAEEvaluate model performance
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)Refine computational models based on the data they have generated.
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- OECD–EC (Europe)Assess an AI system's performance using defined criteria and benchmark tests (Shape 2, intermediate).
- Australia + NZUnderstand how mathematical skills and concepts — pattern recognition, spatial reasoning, chance, data and algorithms — are applied to train AI systems
Improving AI Learning27 lessons· meets 14 requirements inADEKIBCSTA (US)AI4K12 (US)+2
Lessons
- Raj's 70% Problem
- The AI Improvement Framework
- Why Didn't More Photos Fix It?
- Step 1: Diagnose the Problem
- Improvement Comes In Layers
- Identify the Problem Type
- A Few Wrong Labels Can Block Real Progress
- Step 2: Improve Your Data
- Match the Augmentation Technique
- Balancing and Cleaning Data
- Improvement Boost Lab
- Photo Search Runs On Labels
- Step 3: Improve Your Features
- Select the Right Features
- Step 4: Tune Your Model - Hyperparameters
- Tuning Hyperparameters
- Hyperparameter Effects
- Advanced: Ensemble Methods
- Advanced: Transfer Learning
- Match Technique to Situation
- The Improvement Cycle
- Right vs Wrong Approach
- Plan Priya's Fix
- AI Improvement Quiz
- Explain the Improvement Cycle
- Improvement Vocabulary
- AI Improvement Expert!
Requirements this topic helps meet
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)Collect data using computational tools and transform the data to make it more useful and reliable.
- CSTA (US)Discuss issues of bias and accessibility in the design of existing technologies.
- CSTA (US)Describe the ways that bias can be introduced and mitigated in an AI model.
- CSTA (US)Explore strategies to turn ethical considerations into actions, such as mitigating bias in datasets.
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from analyzing a decision-tree-building game and how the learning algorithm adds/replaces nodes (3-5), to comparing how decision-tree learning (add a node) differs from neural-network learning (adjust every weight) (6-8), to describing how decision-tree, neural-net, regression, and reinforcement-learning algorithms each adjust their own type of internal representation (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from demonstrating how weights are assigned in a neural network to produce a desired input/output behavior (3-5), to demonstrating a learning rule (e.g. perceptron rule) that adjusts weights of a one-layer network in response to errors (6-8), to training a multilayer neural network with the backpropagation algorithm and describing how weights and hidden-unit outputs change during learning (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- UNESCOAcquire age-appropriate knowledge and skills on datasets — applying tools/languages to acquire, clean and transform data into a form suitable for storing, processing and analysing (e.g. SQL, NoSQL, SparkSQL, Apache Flink) (CG4.2.3.1).
- Australia + NZStudents are taught that generative AI tools must be used, and their outputs treated, in ways that are accessible, fair and respectful to all users, being alert to potential bias in AI outputs
Unit Review22 lessons· meets 10 requirements inUAEADEKIBCSTA (US)+3
Lessons
- Celebrating Your AI Journey
- Your Unit 2 Journey
- What Rule Connected The Whole Unit?
- Review: ML & Training Data
- You Now Think Like An AI Investigator
- Review: Learning Process
- Review: Testing & Improving
- Match the Core Concepts
- Real AI Teams Review Before They Build Bigger
- Diagnose the Problem
- Hands-On Project Time!
- Your Project, in Miniature
- Project Readiness Check
- Google Translate Has to See Before It Can Translate
- Plan Your Project
- Real-World AI Careers
- What You've Mastered
- Final Concept Check
- Unit 2 Comprehensive Quiz
- Explain What Matters Most
- Unit 2 Key Vocabulary
- Unit 2 Complete - AI Investigator!
Requirements this topic helps meet
- UAEEvaluate model performance
- ADEKExamine dataset bias, propose fixes
- IBExplain fundamental machine learning concepts, including how models are trained on data to make predictions or decisions
- CSTA (US)Refine computational models based on the data they have generated.
- CSTA (US)Describe the roles that humans play (including in data curation and labeling) in creating and refining AI models.
- AI4K12 (US)Progress from identifying patterns/features in labeled data that predict a label (K-2), to modeling how supervised learning encodes discovered patterns as neural-net weights or decision-tree nodes (3-5), to modeling how unsupervised learning finds clusters in unlabeled data (6-8), to modeling how machine learning constructs a reasoner by adjusting a mathematical model's parameters, e.g. linear/polynomial/logistic regression (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- OECD–EC (Europe)Assess an AI system's performance using defined criteria and benchmark tests (Shape 2, intermediate).
- UNESCOUnderstand human roles across the AI life cycle (researchers, architecture/data engineers, beta testers, ethics/safety regulators, human-AI interface specialists, compliance auditors) and the main ethical issues in using data to train AI systems (CG4.1.3.4).
Unit 3: Computer Vision Basics 10 topics
How AI Sees26 lessons· meets 21 requirements inUAEADEKCBSECISCE + states+6
Lessons
- The Missing Sixth Face
- What AI Sees
- What Does It Find Before It Finds A Face?
- Human Vision in Order
- A Face Is In The Arrangement, Not In The Pixels
- Human vs AI
- The Same Face, Almost None Of The Same Numbers
- Camera Sensor
- Pixel Zoom
- Match the Terms
- From Pixels → Patterns
- What Makes A Vision Model More Sure?
- Google Lens Turns Pixels Into An Answer
- Vision Pipeline
- What Is CV?
- AI or Not?
- Who Wins When?
- Spot the AI!
- Global Use Cases
- Use Case → Task
- Three Open Questions
- OCR in Order
- How AI Sees Quiz
- Explain The Five Faces
- Vision Words
- Vision Explorer!
Requirements this topic helps meet
- UAECompare AI with humans (Cycle 2 part)
- ADEKAI vs non-AI decision, AI in daily life
- ADEKSensors / perception
- CBSESummarise basic ideas and concepts of AI and its applications; describe key differences between machine intelligence and human intelligence; explain the difference between automation and AI using real-world cases; differentiate supervised, unsupervised, and reinforcement learning (Class 6).
- CBSEDistinguish key predictive techniques - regression, classification, clustering - and explain key AI domains: Data Science, Computer Vision, Natural Language Processing; explain uses of AI in healthcare, education, transport, communication (Class 7).
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- England + CambridgeExplore Artificial Intelligence (computers taking information from surroundings to produce outputs), machine learning, autonomous robotics, and augmented reality applications
- AI4K12 (US)Progress from identifying the five human senses and sensory organs (K-2), to comparing how different animals perceive the world differently than humans, e.g. sonar, ultraviolet vision, colorblindness (3-5), to giving examples of how humans combine information from multiple sensory modalities to resolve ambiguous signals (6-8); AI4K12 marks 9-12 N/A for this topic since it is considered adequately covered by the lower bands (leaving deeper biology/psychology to other courses).
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from giving examples of intelligent vs. non-intelligent machines and discussing what makes a machine 'intelligent' (K-2), to demonstrating a machine perception tool (e.g. speech transcription, object recognition) and explaining why it is perception rather than mere sensing (3-5), to giving examples of specialized perception algorithms (face detection, obstacle detection, speech recognition, etc.) (6-8), to explaining how perception algorithms power real-world applications like license plate readers and Google Lens (9-12).
- AI4K12 (US)Progress from giving examples of visual features (parts, textures, colors) used to recognize an object class such as cats (K-2), to illustrating how face detection extracts facial features (3-5), to simulating an edge detector to illustrate feature extraction from images (6-8), to explaining how features (formants, edges, intensity gradients) are extracted from waveforms and images at multiple scales (9-12).
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- OECD–EC (Europe)Break a problem into parts and consider AI support for specific steps (Manage 3, intermediate).
- UNESCOUse examples of AI tools (facial recognition, recommenders, self-driving cars, loan-default prediction, etc.) to explain the definition and scope of AI — what AI is and is not — and identify the main categories of AI technologies in daily use (CG4.1.3.1).
- SingaporeDevelop a basic understanding of what AI is, as part of the core Code for Fun curriculum (before taking the AI for Fun elective).
- SingaporeUnderstand what AI is.
- SingaporeLearn about AI: understand what AI is and how it works.
- Korea, Japan, HKExplain the concept and characteristics of artificial intelligence and distinguish AI software from non-AI software ([9정04-01]).
- Korea, Japan, HKUnderstand and experiment hands-on with computer-vision AI functions (object/face recognition, object tracking, colour recognition, line detection, tag recognition) using an AI robotics kit, progressing from basic concepts to a food-delivery/robotics experiment.
Pixels and Pictures26 lessons· meets 10 requirements inCBSECISCE + statesEngland + CambridgeCSTA (US)+2
Lessons
- The Projector Mystery
- Zoom Until It Breaks
- Same Photo, Same Zoom, Different Result
- Pixel and Resolution
- A Picture Is Assembled, Not Stored Whole
- Over A Million?
- Double The Sides, Quadruple The Work
- The Pixel Grid
- A Face In 64 Pixels
- Pixel Terms Match
- Stored As Numbers
- How Many Pixels Does Clear Need?
- Your Gallery App Starts With The Grid
- Rank The Resolutions
- Detail Has A Price
- Pixel Workload
- Edges and Shapes
- Spot The Vision AI!
- Pixels At Work
- Pick The Right Size
- Try It Yourself
- Which File Is Biggest?
- Pixels Quiz
- Explain Pixels and Resolution
- Pixel Words
- Pixel Pro!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- England + CambridgeUnderstand how instructions are stored and executed, and how data of various types (text, sound, pictures) can be represented and manipulated digitally in binary
- CSTA (US)Represent data using multiple encoding schemes.
- CSTA (US)Create and evaluate different abstract representations (e.g., subway map).
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- AI4K12 (US)Progress from constructing a map of a home, school, or neighborhood and understanding that 'the map is not the territory' (K-2), to demonstrating how different map styles (geometric vs. topological) capture different aspects of the world (3-5), to showing how a game board (tic-tac-toe, chess) can be represented as a plain-language description of squares, relationships, and piece positions (6-8), to describing how to represent a concept as a schema.org-style schema with superconcepts and inherited properties (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
Colors for Computers24 lessons· meets 9 requirements inCBSECISCE + statesEngland + CambridgeCSTA (US)+2
Lessons
- RGB Mystery
- Mixing Basics
- Why Can Three Lights Make So Many Colors?
- What Is RGB
- Millions Of Three-Number Recipes, Rewritten Every Frame
- Mix To Match
- Why Do Some Colors Feel Easier To Spot?
- 0–255 Scale
- Color Recipes
- RGB Space
- RGB Mix Lab
- Canva Stores Your Color As Numbers
- Spot The Color Reader!
- Mango Lab
- Hex and RGB
- Make A Color
- RGB Quiz
- Explain RGB Mixing
- Color Words
- Color Detective
- Adds Or Absorbs?
- Hex ↔ RGB Match
- Dominant Channel
- Color Master!
Requirements this topic helps meet
- CBSEInterpret and solve multi-step, multi-layered problems using advanced 3-D transformations, algebraic/multi-rule numerical patterns, interdependent-clue decomposition, and rule-based algorithms with conditional branching - increasing in complexity from Class 6 through Class 8 (e.g., Class 8 adds number-system generalisation across decimal/binary/ternary/Roman/Chinese numerals, and optimisation problems).
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- England + CambridgeUnderstand how instructions are stored and executed, and how data of various types (text, sound, pictures) can be represented and manipulated digitally in binary
- CSTA (US)Represent data using multiple encoding schemes.
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
Finding Edges and Shapes27 lessons· meets 8 requirements inCBSECISCE + statesAI4K12 (US)Singapore+1
Lessons
- Edge Mystery
- Why Do Outlines Stand Out So Fast?
- Big Jumps And Corners
- Edge Basics
- Throwing Almost Everything Away Can Make A Picture Easier To Read
- Edge Types Sort
- Why Are Corners So Helpful?
- Edge Math
- Across Or Down?
- Edge Steps Match
- Edges → Shapes
- Edge Detector Lab
- Scanner Apps Start By Hunting For Four Edges
- Shape Pipeline
- Tricky Edges
- Real Or False Edge?
- Shape Hunt Match
- Edges In Life
- Spot The Edge!
- Quick Recap
- Pipeline Sort
- Edge Tips
- Shape Finder
- Edges & Shapes Quiz
- Explain How Edges Become Shapes
- Edge Words
- Shape Scout!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from giving examples of visual features (parts, textures, colors) used to recognize an object class such as cats (K-2), to illustrating how face detection extracts facial features (3-5), to simulating an edge detector to illustrate feature extraction from images (6-8), to explaining how features (formants, edges, intensity gradients) are extracted from waveforms and images at multiple scales (9-12).
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
- Korea, Japan, HKUse a simulation tool (e.g. a traffic-light decision simulator) to explore how an AI system makes sequential decisions under simulated conditions.
Face Detection Magic26 lessons· meets 10 requirements inUAECBSECISCE + statesCSTA (US)+3
Lessons
- Photo Puzzle
- Detect ≠ Recognize
- How Does A Phone Spot Faces So Quickly?
- Teach The Face Test
- A Layout Can Be A Powerful Signal
- Face-Like? Sort!
- Why Do Cartoons Trigger Face Detection?
- Face Template
- Feature Match
- Detect Steps
- Face Box Lab
- Photo Apps Find The Faces First
- Pipeline Order
- Landmarks 101
- Landmark Sort
- Hard Cases
- Spot The Face!
- Helpful Uses
- Use → Detection
- Modern Detectors
- Align With Points
- Design With Care
- Face Detect Quiz
- Explain Face Detection
- Face Words
- Face Finder!
Requirements this topic helps meet
- UAEHow AI generates outputs from learned patterns
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- OECD–EC (Europe)Be interested in experimenting with various types of AI systems depending on personal needs (e.g. virtual assistants, image analysis, speech/face recognition, robots) (AI example 71, competence 5.2).
- SingaporeLearn about AI: understand what AI is and how it works.
Object Recognition27 lessons· meets 13 requirements inCBSECISCE + statesCSTA (US)AI4K12 (US)+3
Lessons
- Name That Thing
- Edges → Labels
- How Does An App Guess The Right Object?
- Feature → Object
- Features Become Labels Through Learned Patterns
- Classify It!
- Why Does A Split Score Mean 'Not Sure'?
- Train vs Test
- Recognize Order
- Confidence 101
- Recognition Confidence Lab
- Google Lens Combines Features And Confidence
- Confidence Sort
- Many Categories
- Label → Group
- Hard Scenes
- Spot The Object!
- Real Apps
- App → Task
- Ambiguity & Bias
- Fair Training Data?
- Top-1 to Top-5
- Recognition Quiz
- Explain Object Recognition
- Object Words
- Label Practice
- Object Master!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- CSTA (US)Students build data fluency and examine the societal impacts of large-scale data collection used to train AI systems.
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from giving examples of intelligent vs. non-intelligent machines and discussing what makes a machine 'intelligent' (K-2), to demonstrating a machine perception tool (e.g. speech transcription, object recognition) and explaining why it is perception rather than mere sensing (3-5), to giving examples of specialized perception algorithms (face detection, obstacle detection, speech recognition, etc.) (6-8), to explaining how perception algorithms power real-world applications like license plate readers and Google Lens (9-12).
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- AI4K12 (US)Progress from discussing why intelligent agents need to understand languages other than English (K-2), to discussing why domain knowledge must be broad enough to cover accents, dialects, and musical genres (3-5), to describing how a vision system can exhibit cultural bias if it lacks knowledge of objects outside its creators' culture, e.g. non-US traffic signs (6-8), to describing technical difficulties (e.g. low-contrast faces, children's speech) in making perception systems work well for diverse groups (9-12).
- AI4K12 (US)Progress from evaluating the ways a decision impacts different people differently (K-2), to evaluating how an AI technology, e.g. speech or facial recognition, has different effects on different demographic groups (3-5), to evaluating how various stakeholders' conflicting goals and values shape AI system design decisions (6-8), to explaining how the use of AI systems (e.g. health-risk algorithms) has led to documented disparate impacts on different groups (9-12).
- OECD–EC (Europe)Compare the importance of data selection and collection methods for an AI system (Shape 3, intermediate).
- SingaporeLearn about AI: understand what AI is and how it works.
- Korea, Japan, HKUnderstand and experiment hands-on with computer-vision AI functions (object/face recognition, object tracking, colour recognition, line detection, tag recognition) using an AI robotics kit, progressing from basic concepts to a food-delivery/robotics experiment.
AI Reads Text24 lessons· meets 7 requirements inCBSECISCE + statesAI4K12 (US)Singapore
Lessons
- Priya's Grandfather's Book
- What OCR Really Does
- How Can A Camera Turn A Photo Into Editable Words?
- The 5 Steps of OCR
- OCR Turns Visual Patterns Into Language
- Order the Steps!
- Why Is Messy Text So Hard To Read?
- OCR in Action - True or False?
- Match the OCR Challenge
- Easy or Hard for OCR?
- Scripts and Their Puzzles
- OCR Accuracy Lab
- Google Lens Reads Text Straight From Your Camera
- Script Challenge Match!
- Where OCR Is Really Used
- OCR Apps Match-Up!
- How Much Can You Trust It?
- Sort by Trust!
- Spot the OCR!
- What Is Actually Getting Better
- OCR Master Quiz!
- Explain How OCR Reads Text
- OCR Vocabulary
- OCR Champion!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- AI4K12 (US)Progress from discussing why intelligent agents need to understand languages other than English (K-2), to discussing why domain knowledge must be broad enough to cover accents, dialects, and musical genres (3-5), to describing how a vision system can exhibit cultural bias if it lacks knowledge of objects outside its creators' culture, e.g. non-US traffic signs (6-8), to describing technical difficulties (e.g. low-contrast faces, children's speech) in making perception systems work well for diverse groups (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
Image Classification25 lessons· meets 10 requirements inCBSECISCE + statesCSTA (US)AI4K12 (US)+1
Lessons
- Raj's Photo Chaos
- Which Job Gives Which Answer?
- How Can One Photo Fit More Than One Category?
- Single vs Multi-Label
- Categories Are Learned, Not Hard-Coded
- Single or Multi-Label?
- Why Look-Alikes Split the Score
- The 5 Steps of Classification
- Order the Steps!
- Now Run the Five Steps Yourself
- Can More Photos Fix It?
- The Cut-Off Lab
- Google Photos Uses Multi-Label Classification
- Easy or Hard to Classify?
- Classification Everywhere!
- Classification Apps Match!
- Planning a Street Food Classifier
- Food Descriptions Match!
- True, or Hype?
- Spot the Classification!
- The Future of Classification!
- Classification Master Quiz!
- Explain Image Classification
- Classification Vocabulary
- Classification Champion!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from giving examples of visual features (parts, textures, colors) used to recognize an object class such as cats (K-2), to illustrating how face detection extracts facial features (3-5), to simulating an edge detector to illustrate feature extraction from images (6-8), to explaining how features (formants, edges, intensity gradients) are extracted from waveforms and images at multiple scales (9-12).
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
Computer Vision in the Real World44 lessons· meets 30 requirements inUAEADEKCBSECISCE + states+7
Lessons
- Priya's Travel-Day Stakeout
- Priya's Tally Gets Complicated
- Why Does Computer Vision Show Up In So Many Places?
- Biometric ID: Which Part Is Vision?
- How Biometric Enrollment Works
- One Vision Pipeline Can Power Many Industries
- Order the Verification Steps
- Why Do Sectors Need Different Vision Rules?
- Where is Biometric CV Used?
- Paperless Airport Boarding
- Order the 5 Jobs Behind the Gate
- Deployment Trade-Off Lab
- DigiYatra Connects Several CV Steps
- Sort the Benefits
- Smart Traffic Monitoring
- Smart City CV Applications
- CV in Railways & Transportation
- CV in Agriculture & Farming
- CV in Healthcare & Medicine
- CV in Shopping & E-Commerce
- CV in Education
- CV in Banking & Finance
- CV in Heritage & Preservation
- Sort CV Applications by Sector
- Reading CV Numbers Like an Investigator
- Why CV Spreads - And Where It Stalls
- Investigation Challenge 1: CV in Your Life
- Investigation Challenge 2: Privacy vs. Convenience
- Investigation Challenge 3: Future Applications
- Match CV Types to Applications
- Spot the CV - Real World Edition
- CV Applications Quiz - Part 1
- CV Applications Quiz - Part 2
- Explain Real-World Computer Vision
- Key Terms
- Key Takeaways
- Byte's Final Message
- Case Study: The School Gate That Never Sleeps
- Write the Rule for the School Gate
- Case File: The Score in the Courtroom
- Who Has Something at Stake?
- Fill In the Ethical Matrix
- Take a Position on RiskCheck
- Chapter Complete!
Requirements this topic helps meet
- UAEPrivacy, accountability and personal rights
- UAEHow AI is governed through national policies
- UAEAI applied across sectors
- ADEKStakeholder debate / ethical matrix
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- England + CambridgeUnderstand data security during transmission and different user authentication methods (password, fingerprint, facial recognition)
- England + CambridgeUnderstand that robots can work autonomously and describe robotics benefits in manufacturing/food production
- CSTA (US)Apply multiple methods of encryption to model the secure transmission of information.
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- CSTA (US)Identify the intended and unintended impacts of AI on society, including government, education, entertainment, culture, careers, and national security, while considering how these impacts may differ among diverse communities.
- AI4K12 (US)Progress from identifying the five human senses and sensory organs (K-2), to comparing how different animals perceive the world differently than humans, e.g. sonar, ultraviolet vision, colorblindness (3-5), to giving examples of how humans combine information from multiple sensory modalities to resolve ambiguous signals (6-8); AI4K12 marks 9-12 N/A for this topic since it is considered adequately covered by the lower bands (leaving deeper biology/psychology to other courses).
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- AI4K12 (US)Progress from identifying the features that make objects in a collection unique and organizing them into a feature table (K-2), to constructing a feature-vector representation for a set of objects and showing that similar objects lie close together in feature space (3-5), to explaining how word embeddings represent words as numeric vectors for tasks like machine translation (6-8), to describing how a transformer network maps sequences of word-embedding vectors to output sequences (9-12).
- AI4K12 (US)Progress from evaluating the ways a decision impacts different people differently (K-2), to evaluating how an AI technology, e.g. speech or facial recognition, has different effects on different demographic groups (3-5), to evaluating how various stakeholders' conflicting goals and values shape AI system design decisions (6-8), to explaining how the use of AI systems (e.g. health-risk algorithms) has led to documented disparate impacts on different groups (9-12).
- AI4K12 (US)Progress from discussing what makes a system fair vs. unfair through a fair-distribution exercise (K-2), to evaluating whether an AI system meets the design criteria of transparency and explainability (3-5), to evaluating whether an AI system meets the design criteria of accountability and respect for privacy (6-8), to analyzing whether an AI system satisfies the full set of ethical design criteria: fairness, transparency, explainability, accountability, respect for privacy, and adherence to societal values (9-12).
- AI4K12 (US)Progress from analyzing deepfake images/videos to identify the flaws that currently reveal them, e.g. thispersondoesnotexist.com (3-5), to critiquing uses of AI technology for surveillance or privacy violation, e.g. face recognition, license-plate trackers (6-8), to identifying appropriate areas for AI regulation and evaluating proposed legislation, e.g. autonomous-vehicle laws (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Identify unintended impacts of an AI use that differ from its immediate goals (Engage 7, intermediate).
- OECD–EC (Europe)Propose specific design changes to improve an AI system for human well-being (Shape 4, intermediate).
- UNESCOIllustrate dilemmas around AI and identify the main reasons behind ethical conflicts that AI creators and companies face (e.g. data collection vs. ownership, profit vs. privacy, safety vs. speed of iteration) (CG4.1.2.1).
- UNESCODiscuss age-appropriate real-world cases around six core AI ethical principles — do no harm, proportionality, non-discrimination, sustainability, human determination, and transparency/explainability — and their personal implications (CG4.1.2.2).
- UNESCOInternalise ethical principles on AI (human rights, data privacy, human agency, equity, inclusion, social justice, environmental sustainability) into reflective practices, e.g. advocating for inclusive AI interfaces and reporting discriminatory bias found in AI tools (CG4.1.2.3).
- UNESCOEvaluate the consequences of institutional use of AI to make high-stakes decisions about people (e.g. profiling students, hiring, loan/assessment scoring) and understand why human accountability must not be ceded to AI in such decisions (CG4.2.1.2).
- SingaporeLearn about AI: understand what AI is and how it works.
- Korea, Japan, HKIdentify AI ethical principles and analyse ethical issues raised by real AI applications (bias, fairness, accountability), as a dedicated unit distinct from general 'societal impact'.
- Korea, Japan, HKUnderstand and experiment hands-on with computer-vision AI functions (object/face recognition, object tracking, colour recognition, line detection, tag recognition) using an AI robotics kit, progressing from basic concepts to a food-delivery/robotics experiment.
- Korea, Japan, HKEvaluate the societal impact of AI, including both opportunities and risks it poses to communities, drawing on real cases (e.g. AlphaFold's protein-folding breakthrough, AI-assisted astronomy, AI-generated music).
Unit 3 Review & Project38 lessons· meets 8 requirements inCBSECISCE + statesAI4K12 (US)Singapore+1
Lessons
- Priya's Tally, One Unit Later
- Review: Chapters 21-24
- What Do All Nine Chapters Have In Common?
- Review: Chapters 25-29
- The CV Pipeline
- A Few Core Ideas Power A Great Many Vision Tasks
- Order the CV Pipeline
- Why Real Products Chain Several Vision Tasks
- The Building Blocks of CV
- Match Building Blocks
- Types of CV Tasks
- Project Design Lab
- One Camera View, Four Of Your Five Tasks
- Match CV Task Types
- Unit 3 Quiz - Part 1
- Unit 3 Quiz - Part 2
- Unit 3 Quiz - Part 3
- Explain The Pipeline In Your Own Words
- Sort CV Applications
- Think Deeper - Big Questions
- Project Part 1: CV Detective Work
- Project Part 2: Train One Yourself
- Project Part 3: Design Your Own CV Solution
- Build a Classifier: Five Steps
- Unplugged Activities (No Computer Needed!)
- CV Careers - Jobs of the Future
- Match CV Careers
- Skills That Do the Work
- Benefit, Risk or Open Question?
- Byte's Celebration Message
- What's Next: Unit 4 Preview
- Unit 3 Key Terms Review
- Final Challenge: Master CV Spotter
- Who Makes the Final Call?
- Certificate-Worthy!
- Final Reflection
- How Your Project Is Judged
- Unit 3 Complete!
Requirements this topic helps meet
- CBSEExplain computer vision and its significance in visual data analysis; understand key stages (acquisition, preprocessing, feature extraction, analysis); identify real-world applications in healthcare, surveillance, autonomous vehicles; analyse challenges (ethics, privacy, technical limitations); develop basic skills using OpenCV and deploying ML models online (Unit 3).
- CISCE + statesIntroduce Computer Vision and OpenCV applications in real-world scenarios: image classification, object detection, facial recognition, colour detection and object tracking (brief explanation with examples)
- AI4K12 (US)Progress from locating and identifying computer sensors such as cameras and microphones on everyday devices (K-2), to illustrating how computer sensing differs from human sensing (3-5), to giving examples of intelligent agents (e.g. self-driving cars) combining data from multiple sensors (6-8), to describing the limitations and advantages of specific sensor types such as cameras, microphones, radar, lidar, GPS, and accelerometers (9-12).
- AI4K12 (US)Progress from explaining how images are digitally represented as 2D pixel arrays (3-5), to explaining how sounds are digitally encoded via waveform sampling (6-8), to explaining how radar, lidar, GPS, and accelerometer data are represented and computed (9-12). K-2 is marked N/A for this topic.
- AI4K12 (US)Progress from giving examples of visual features (parts, textures, colors) used to recognize an object class such as cats (K-2), to illustrating how face detection extracts facial features (3-5), to simulating an edge detector to illustrate feature extraction from images (6-8), to explaining how features (formants, edges, intensity gradients) are extracted from waveforms and images at multiple scales (9-12).
- AI4K12 (US)Progress from demonstrating figure/ground segmentation in an image (K-2), to illustrating how occlusion changes the outlines of objects (3-5), to describing how edge detectors compose into more complex feature detectors such as letter/shape detectors (6-8), to demonstrating how perceptual reasoning at a higher abstraction level (objects, surfaces, boundaries) draws on lower levels (contours, edges, pixels) (9-12).
- SingaporeLearn about AI: understand what AI is and how it works.
- Korea, Japan, HKUnderstand and experiment hands-on with computer-vision AI functions (object/face recognition, object tracking, colour recognition, line detection, tag recognition) using an AI robotics kit, progressing from basic concepts to a food-delivery/robotics experiment.
Unit 4: Language and AI 10 topics
AI Understands Words35 lessons· meets 18 requirements inUAEADEKCBSECISCE + states+5
Lessons
- The Sentence Nani Would Not Say
- The Big Question
- How Did The Phone Handle Nani's Own Words?
- Three Jobs, Not One
- The Wording Changed And The Meaning Survived
- The Word Journey
- Why Can One Sentence Mean Different Things?
- Follow The Chain
- Three Layers of Language
- Did The Order Change The Meaning?
- Which Meaning Is In Play?
- Layer 3: Context
- Sort The Three Layers
- The 4 NLP Steps
- Intent Confidence Lab
- One Spoken Question, Two Different AI Jobs
- Count The Tokens
- What Job Is Each Word Doing?
- Step 3: Meaning
- Look It Up, Or Write It Out?
- Order the Steps
- Intent Matching
- NLP All Around
- NLP Apps Match
- Four Tricky Sentences
- Tricky Sentence Sort
- World Languages
- Spot the NLP!
- NLP Knowledge Quiz
- Explain What Happened To Nani's Question
- NLP Vocabulary
- Nani's Question, End To End
- Byte's Message
- Next Chapter Preview
- Chapter Complete!
Requirements this topic helps meet
- UAEHow AI generates outputs from learned patterns
- ADEKClassification vs generation, how generative AI is trained
- CBSEDefine Generative AI and classify different kinds; explain how Generative AI works and how it learns, differentiate from conventional AI; apply Generative AI tools to create content; understand ethical considerations of using Generative AI (Unit 4: Introduction to Generative AI).
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- AI4K12 (US)Progress from describing the sounds that make up spoken language and giving example words for each vowel sound (K-2), to illustrating how sequences of sounds are recognized as candidate words despite unclear signals (3-5), to illustrating how sequences of words are recognized as phrases by resolving ambiguity with higher-level information (6-8), to illustrating the full abstraction hierarchy from waveform to sentence in speech understanding (9-12).
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- AI4K12 (US)Progress from demonstrating knowledge of language structure by generating plausible/implausible novel words or unscrambling sentences (K-2), to demonstrating that human language is infinite via sentence extension (3-5), to using a parser to display a sentence's syntactic (parse-tree) structure and explaining the node categories (6-8), to demonstrating how a small context-free grammar can parse or generate simple sentences, and its limits with agreement constraints (9-12).
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating how a computer produces different verb tense/agreement forms (K-2), to experimenting with a speech-to-text system's ability to resolve ambiguous words via context (3-5), to illustrating how word embeddings support reasoning about word meaning, e.g. outlier detection and analogy completion (6-8), to demonstrating how a small context-free grammar parses or generates simple sentences and where it falls short, e.g. subject-verb agreement (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- AI4K12 (US)Progress from giving commonsense explanations of human behavior useful for a robot to know, e.g. choosing a plausible story ending (K-2), to explaining what everyday cultural and naive-physics knowledge a computer needs to understand a story (3-5), to explaining the naive-physics inference rules (mass, volume, conservation, forces) needed to solve a physics reasoning problem (6-8), to explaining the cultural, naive-physics, and folk-psychology knowledge needed to interpret a fable, fairytale, or Winograd-schema sentence (9-12).
- AI4K12 (US)Progress from demonstrating ways intelligent agents don't understand things the way people do, e.g. no frustration at repeated questions (K-2), to demonstrating what a 'broad AI'/general-AI system could do that current narrow AI cannot (3-5), to defining criteria for consciousness (self-awareness, personal history) and evaluating AI systems or fictional AI characters against them (6-8), to debating alternative perspectives on human vs. artificial intelligence, including the Turing test and artificial superintelligence (9-12).
- OECD–EC (Europe)Use AI tools within a guided national learning app designed to build critical thinking by prompting students to find answers and form their own conclusions rather than supplying direct answers.
- SingaporeLeverage AI tools effectively for learning and for daily life.
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- ChinaStudents understand the technical characteristics of generative AI and its significance to national strategy
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
How Chatbots Work33 lessons· meets 11 requirements inADEKCISCE + statesEngland + CambridgeCSTA (US)+4
Lessons
- The Mystery Bot
- Raj's Big Question
- How Does A Bot Always Ask The Next Right Question?
- Two Doors Into a Chatbot
- A Strict Flowchart Can Still Feel Like A Conversation
- Two Types of Bots
- Why Do Small Wording Changes Confuse Some Bots?
- Which Question Comes Next?
- Pizza Bot Example
- The Other Branch
- 4 Key Components
- Bot Helpfulness Lab
- Ordering Bots Live Or Die By Their Flow
- Train an Intent Yourself
- Keywords & Patterns
- Entities
- Written in Advance, or Dropped In?
- Component Match
- Library Bot Demo
- The Bot's Rules
- The Renew Branch
- 4 Bot Problems
- Problem Sorting
- Rules or AI? Real Bots
- 4 Bot Upgrades
- Upgrade Match
- Spot the Chatbot!
- Chatbot Quiz
- Explain How A Rule-Based Chatbot Works
- Chatbot Vocabulary
- Run the Whole Bot
- Byte's Message
- Chapter Complete!
Requirements this topic helps meet
- ADEKChatbot mechanics, build a simple conversational agent
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- England + CambridgeFollow and create algorithms presented as flowcharts using standard symbols (start, stop, process, decision, connector), including multiple variables
- England + CambridgeFollow, edit and create algorithms using flowcharts and pseudocode; use selection statements and Boolean logic; progress from linear to binary search
- CSTA (US)Use flowcharts and/or pseudocode to address complex problems as algorithms.
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- OECD–EC (Europe)Be aware that some AI systems aim to provide human-like interaction (e.g. customer-service chatbots) and be able to interact with and give feedback to such a system to influence what it recommends next (AI examples 8 and 33, competence 2.1).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
Family Reunion33 lessons· meets 12 requirements inUAECISCE + statesAI4K12 (US)Singapore+2
Lessons
- Family Reunion
- It Works!
- Dadi's Saying
- Why Did Dadi's Saying Break The App?
- Name The Job
- A Sentence Can Change Language And Keep Its Idea
- Rules Or Examples?
- Why Do Some Phrases Break In Translation?
- Parallel Corpus
- Match Translations
- Which Bat?
- Encode & Decode
- Translation Accuracy Lab
- Real Apps Learn From Enormous Piles Of Translated Text
- Build Pipeline
- Attention Focus
- Which Pairs Work Best
- Sort The Pairs
- Idiom Problems
- No Exact Match
- Tum Or Aap?
- Gender Challenge
- Word Order
- Match The Problems
- How It Is Built
- Translate Features
- Match Features
- Code-Switching
- Spot Translation AI
- Translation Quiz
- Explain How A Translation Model Carries Meaning
- New Words
- Chapter Complete!
Requirements this topic helps meet
- UAEEthics, cultural values and inclusion guide AI decisions
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- AI4K12 (US)Progress from demonstrating knowledge of language structure by generating plausible/implausible novel words or unscrambling sentences (K-2), to demonstrating that human language is infinite via sentence extension (3-5), to using a parser to display a sentence's syntactic (parse-tree) structure and explaining the node categories (6-8), to demonstrating how a small context-free grammar can parse or generate simple sentences, and its limits with agreement constraints (9-12).
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating how a computer produces different verb tense/agreement forms (K-2), to experimenting with a speech-to-text system's ability to resolve ambiguous words via context (3-5), to illustrating how word embeddings support reasoning about word meaning, e.g. outlier detection and analogy completion (6-8), to demonstrating how a small context-free grammar parses or generates simple sentences and where it falls short, e.g. subject-verb agreement (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- AI4K12 (US)Progress from giving commonsense explanations of human behavior useful for a robot to know, e.g. choosing a plausible story ending (K-2), to explaining what everyday cultural and naive-physics knowledge a computer needs to understand a story (3-5), to explaining the naive-physics inference rules (mass, volume, conservation, forces) needed to solve a physics reasoning problem (6-8), to explaining the cultural, naive-physics, and folk-psychology knowledge needed to interpret a fable, fairytale, or Winograd-schema sentence (9-12).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Korea, Japan, HKUnderstand speech recognition and natural-language-processing techniques and try a word-association-style generative-AI tool to see how language models generate text.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
- Australia + NZStudents are taught that generative AI tools must be used, and their outputs treated, in ways that are accessible, fair and respectful to all users, being alert to potential bias in AI outputs
Autocorrect and Prediction30 lessons· meets 7 requirements inCISCE + statesAI4K12 (US)SingaporeAustralia + NZ
Lessons
- Embarrassing Message
- Meat vs Meet
- Real Word, Or Not?
- Why Did The Keyboard Leave 'Meat' Alone?
- Autocorrect Or Spell Check?
- Every Word You Type Sets Off A Quiet Guess
- How Typing Help Grew Up
- Why Can One Typo Have Two Different Fixes?
- Step 1: The Word List
- One Change, Or More?
- Step 3: Probability
- The Neighbors Decide
- Fix It Or Leave It
- Autocorrect Confidence Lab
- Gboard Runs All Four Steps On Your Phone
- Predictive Text
- N-gram Patterns
- Match N-grams
- Teach It Your Own Examples
- Why Did It Fail?
- When The Keyboard Overrules You
- Where The Keyboard Learns From
- Rebuild The Swipe
- Smart Reply
- Match Replies
- Spot The Prediction
- Autocorrect Quiz
- Explain How A Keyboard Chooses A Fix
- New Words
- Chapter Complete!
Requirements this topic helps meet
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating how a computer produces different verb tense/agreement forms (K-2), to experimenting with a speech-to-text system's ability to resolve ambiguous words via context (3-5), to illustrating how word embeddings support reasoning about word meaning, e.g. outlier detection and analogy completion (6-8), to demonstrating how a small context-free grammar parses or generates simple sentences and where it falls short, e.g. subject-verb agreement (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
Sentiment Analysis35 lessons· meets 10 requirements inUAECISCE + statesEngland + CambridgeAI4K12 (US)+3
Lessons
- The Number Nobody Typed
- AI Sorts the Reviews
- Your Turn: Label These Reviews
- How Can Software Guess A Feeling From Words?
- Name That Job
- Three Words Can Carry A Whole Opinion
- Opinion, or Just a Fact?
- Method 1: A Word List
- Find the Opinion Word
- One Tiny Word Can Flip The Whole Mood
- Method 2: Scoring the Words
- Score the Sentence
- Method 3: Learning From Examples
- Sentiment Accuracy Lab
- A Popular Video Gets More Comments Than Anyone Can Read
- Use 1: Listening to Social Media
- Use 2: Review Summaries
- Use 3: Sorting the Support Queue
- Use 4: Public Opinion — and Its Big Trap
- Use 5: Wellbeing Apps — Handle With Care
- Emotion Detection
- Aspect-Based Sentiment
- Intensity: How Strong Is It?
- Challenge 1: Sarcasm
- Challenge 2: Slang Keeps Moving
- Challenge 3: Negation
- Challenge 4: Comparisons
- Challenge 5: Every Language Has Its Own
- Challenge 6: Mixed Opinions
- What Companies Do With It
- Spot the Sentiment AI
- Chapter Quiz
- Explain It In Your Own Words
- Vocabulary Builder
- Chapter Complete!
Requirements this topic helps meet
- UAEAI-manipulated media: safety and security
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- England + CambridgeMedia literacy (taught mainly via Citizenship) should cover critically engaging with messages conveyed through different media channels, including AI, and identifying/protecting against misinformation and disinformation
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- AI4K12 (US)Progress from demonstrating how computers recognize emotions in faces via mouth/eye/eyebrow shape (K-2), to illustrating how sentiment analysis judges the emotional tone of text, e.g. movie/restaurant reviews, social media (3-5), to describing how computers combine facial expression, gaze, gesture, body language, tone of voice, and word choice to recognize emotional states (6-8), to identifying ways AI applications modify their behavior in response to detected human emotional states, e.g. adaptive tutoring systems (9-12).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Korea, Japan, HKUnderstand speech recognition and natural-language-processing techniques and try a word-association-style generative-AI tool to see how language models generate text.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
Voice Assistants Explained30 lessons· meets 11 requirements inUAECISCE + statesCSTA (US)AI4K12 (US)+4
Lessons
- The Three-Speaker Test
- Design a Test That Splits Them
- The Favorite Song Test
- What Happens Between The Microphone And The Music?
- True or Myth: What a Voice Assistant Is
- One Sentence Sets Off A Relay Of Different AI Systems
- Five Assistants, One Chain
- Stage 1: Wake Word Detection
- Stage 2: Speech Recognition
- Why Do Voice Assistants Sometimes Mishear Us?
- Stage 3: Words Into a Request
- Stage 4: Taking Action
- Stage 5: Build the Voice
- What Actually Makes An Assistant Get It Right?
- The Same Sentence, A Different Answer
- The Complete Journey
- Language Support
- Code-Switching
- What They Are Actually For
- Personalization
- Context Awareness
- What Happens to Your Voice
- Privacy Protection
- Name the Stage That Broke
- The Future of Voice Assistants
- Spot the AI: Voice Assistants
- Chapter Quiz
- Explain How A Voice Assistant Answers You
- Vocabulary Builder
- Chapter Complete!
Requirements this topic helps meet
- UAEHow AI generates outputs from learned patterns
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- CSTA (US)Apply multiple methods of encryption to model the secure transmission of information.
- AI4K12 (US)Progress from describing the sounds that make up spoken language and giving example words for each vowel sound (K-2), to illustrating how sequences of sounds are recognized as candidate words despite unclear signals (3-5), to illustrating how sequences of words are recognized as phrases by resolving ambiguity with higher-level information (6-8), to illustrating the full abstraction hierarchy from waveform to sentence in speech understanding (9-12).
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- OECD–EC (Europe)Be interested in experimenting with various types of AI systems depending on personal needs (e.g. virtual assistants, image analysis, speech/face recognition, robots) (AI example 71, competence 5.2).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Korea, Japan, HKUnderstand speech recognition and natural-language-processing techniques and try a word-association-style generative-AI tool to see how language models generate text.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
AI and World Languages31 lessons· meets 15 requirements inUAECISCE + statesCSTA (US)AI4K12 (US)+5
Lessons
- Nani, Weeks Later
- What the Phone Actually Does
- The Sentence That Broke It
- Why Does AI Struggle When Languages Mix Together?
- Thousands of Languages, Two Routes
- Human Language Is Wildly Diverse Yet Still Learnable
- World Writing Scripts
- Challenge 1: The Data Gap
- Challenge 2: Transliteration
- Why Can One Word Have So Many Spellings?
- Challenge 3: Mixing Languages
- Challenge 4: Dialects and Varieties
- Whose Voice Does It Hear?
- Whose Voice Check
- Three More Language Challenges
- Multilingual Support Lab
- Gboard Helps Users Type Across Scripts
- Challenge Matching Game
- Closing the Gap
- Google's Language Support
- Who Is Building Language AI
- Teach the Keyboard Two Scripts
- Language AI in Everyday Apps
- What Researchers Are Working On
- What You Can Actually Do
- Challenge vs Solution
- Spot the Language AI
- Chapter Quiz
- Explain Why Multilingual AI Is Hard
- Vocabulary Builder
- Chapter Complete!
Requirements this topic helps meet
- UAEEthics, cultural values and inclusion guide AI decisions
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- CSTA (US)Students build data fluency and examine the societal impacts of large-scale data collection used to train AI systems.
- AI4K12 (US)Progress from describing what an intelligent agent must 'know' to make sense of a spoken question (K-2), to demonstrating how a text-to-speech/speech-recognition system resolves ambiguity using statistical context (3-5), to classifying an image scene type and describing the domain knowledge needed to understand it (6-8), to analyzing an online image dataset (e.g. ImageNet, Coco) to describe how statistical domain knowledge is derived for a computer vision system (9-12).
- AI4K12 (US)Progress from discussing why intelligent agents need to understand languages other than English (K-2), to discussing why domain knowledge must be broad enough to cover accents, dialects, and musical genres (3-5), to describing how a vision system can exhibit cultural bias if it lacks knowledge of objects outside its creators' culture, e.g. non-US traffic signs (6-8), to describing technical difficulties (e.g. low-contrast faces, children's speech) in making perception systems work well for diverse groups (9-12).
- AI4K12 (US)Progress from demonstrating knowledge of language structure by generating plausible/implausible novel words or unscrambling sentences (K-2), to demonstrating that human language is infinite via sentence extension (3-5), to using a parser to display a sentence's syntactic (parse-tree) structure and explaining the node categories (6-8), to demonstrating how a small context-free grammar can parse or generate simple sentences, and its limits with agreement constraints (9-12).
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating how a computer produces different verb tense/agreement forms (K-2), to experimenting with a speech-to-text system's ability to resolve ambiguous words via context (3-5), to illustrating how word embeddings support reasoning about word meaning, e.g. outlier detection and analogy completion (6-8), to demonstrating how a small context-free grammar parses or generates simple sentences and where it falls short, e.g. subject-verb agreement (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- OECD–EC (Europe)Compare the importance of data selection and collection methods for an AI system (Shape 3, intermediate).
- UNESCOCategorise general safety risks of AI and typical AI incidents, deepen knowledge of human rights to data protection and privacy and creators' legal responsibility to collect data only with informed consent, and practise strategies to ensure personal data is collected, used, shared, archived and deleted only with consent (CG4.2.2.3).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Korea, Japan, HKExamine problems that can arise in the collection and use of data needed for AI learning (bias, privacy, consent) and propose legally, socially and ethically valid solutions ([9정04-05]).
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
Text to Speech30 lessons· meets 10 requirements inUAECISCE + statesCSTA (US)AI4K12 (US)+2
Lessons
- The Long Journey
- AI Reads Stories
- Bed or Seed?
- How Does A Silent Device Suddenly Sound Human?
- What is Text-to-Speech?
- A Voice Can Be Built From Patterns Instead Of A Person
- Recorded, or Built On The Spot?
- Step 1: Text Analysis
- Step 2: Say It Out Loud
- Why Are Numbers And Symbols So Hard To Say Naturally?
- Step 3: Phonetic Conversion
- Step 4: Prosody Generation
- Step 5: Stitched, or Generated?
- Natural Voice Lab
- Google Maps Uses TTS To Speak Directions Clearly
- TTS Process Order
- Evolution of TTS
- Challenge 1: Spot the Trap
- Challenge 2: One Number, Two Ways
- Challenge 3: Emphasis
- Challenge Matching
- Label the Pitch
- Who TTS Is Really For
- Voice Cloning Technology
- Already Here, Or Still Hard?
- Spot the TTS
- Chapter Quiz
- Explain How TTS Turns Text Into A Voice
- Vocabulary Builder
- Chapter Complete!
Requirements this topic helps meet
- UAEAI-manipulated media: safety and security
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- CSTA (US)Discuss issues of bias and accessibility in the design of existing technologies.
- AI4K12 (US)Progress from describing the sounds that make up spoken language and giving example words for each vowel sound (K-2), to illustrating how sequences of sounds are recognized as candidate words despite unclear signals (3-5), to illustrating how sequences of words are recognized as phrases by resolving ambiguity with higher-level information (6-8), to illustrating the full abstraction hierarchy from waveform to sentence in speech understanding (9-12).
- AI4K12 (US)Progress from demonstrating knowledge of language structure by generating plausible/implausible novel words or unscrambling sentences (K-2), to demonstrating that human language is infinite via sentence extension (3-5), to using a parser to display a sentence's syntactic (parse-tree) structure and explaining the node categories (6-8), to demonstrating how a small context-free grammar can parse or generate simple sentences, and its limits with agreement constraints (9-12).
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
AI Writing Stories39 lessons· meets 29 requirements inUAEADEKCISCE + statesCSTA (US)+6
Lessons
- The Writing Challenge
- Team 1: The Clock Tower
- What Gave It Away?
- Why Didn't the Better-Written Half Win?
- Two Kinds of AI
- Two People, One Prompt, Two Different Stories
- Famous AI Writers
- One Machine, Very Different Jobs
- Be the Language Model
- How AI Learns to Write
- Word by Word Generation
- Build It the Model's Way
- Why It Reads So Well
- Story Quality Lab
- ChatGPT Helps Draft And Rewrite Text
- AI Writing Superpowers
- Match the Superpower
- What It Can and Cannot Judge
- No Life to Draw On
- AI Hallucinations
- More AI Limitations
- AI Can vs Can't
- When It's OK to Use AI
- Fair, or Over the Line?
- The Right Working Order
- Brainstorm, Then Merge
- Future of AI Writing
- Spot the AI Writer!
- From Words to Pictures
- Static to Picture
- Strong Check or Weak Hint?
- Who Made It? The Art Show Question
- Your Verdict: The Art Show
- Pictures Check
- AI Writing Quiz
- Explain How AI Generates Story Text
- AI Writing Words
- Key Takeaways
- AI Writing Champion!
Requirements this topic helps meet
- UAECompare AI with humans (Cycle 2 part)
- UAEAI-manipulated media: safety and security
- UAEUse generative AI models to create new content
- ADEKExplain why generative AI can fabricate/hallucinate
- ADEKCheck AI output before trusting
- ADEKHuman/AI-generated art + ethics
- ADEKGenerative image tools (GAN/diffusion), deepfakes
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- CSTA (US)Explore the tradeoffs related to human agency (including privacy, safety, creativity, autonomy, and intellectual property) when AI is used.
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- OECD–EC (Europe)Verify AI outputs against trusted sources (Engage 3, intermediate).
- OECD–EC (Europe)Compare AI-generated suggestions to one's own ideas during brainstorming (Create 1, intermediate).
- OECD–EC (Europe)Consider when attribution, permission, or avoiding AI use is the appropriate choice (Create 4, intermediate).
- OECD–EC (Europe)Be aware that AI can help edit and process digital content (e.g. photo aging, text auto-suggestion), and know how to incorporate AI-edited or AI-generated content into one's own work while considering questions of credit (AI examples 7 and 38, competence 3.2).
- UNESCOExplain that AI is human-led and that decisions by AI creators shape how AI systems affect human rights, human-AI interaction, and society; recognise that AI cannot replace human thinking or intellectual development (CG4.1.1.1).
- UNESCOCritically examine the dynamic relationship between human agency and machine agency, including cases where a degree of machine agency may be warranted (e.g. medical pattern detection) versus cases requiring human control (CG4.1.1.3).
- SingaporeLeverage AI tools effectively for learning and for daily life.
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- SingaporeLearn beyond AI: develop distinctly human capabilities and dispositions that AI cannot replicate.
- SingaporeThink critically about AI's biases, question AI-generated outputs, and discern or identify false or misleading information produced by AI.
- ChinaStudents deepen ethical understanding and critically identify misinformation/hallucination risks in generative-AI outputs
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
- Australia + NZStudents develop an understanding of how generative AI tools work in broad terms, how they can be used appropriately, and when/how such tools are affecting the content and information they see (transparency about AI-generated or AI-assisted content)
- Australia + NZStudents are taught that generative AI tools must be used, and their outputs treated, in ways that are accessible, fair and respectful to all users, being alert to potential bias in AI outputs
- Australia + NZStudents learn that use of generative AI outputs (e.g. in assessable work) must remain open to challenge, retain human agency, and that the student/user remains accountable for decisions made with AI assistance (e.g. academic integrity implications)
Unit 4 Review34 lessons· meets 14 requirements inADEKCBSECISCE + statesIB+6
Lessons
- Welcome Back, Investigators!
- Our Language AI Journey
- Which Kind Of Bot?
- How Do So Many Language AI Tools Use Similar Ideas?
- Review: Translation & Text
- Every Language Tool Breaks In The Same Place
- Rebuild The Voice Assistant
- Use It Or Check It?
- Master Vocabulary Match
- The Words You Just Matched Are The Words Developers Actually Use
- Name The AI In Your Day
- Think: Privacy vs Convenience
- Help, Or Handover?
- Chatbot Design Project
- Step 1: Define Purpose
- Intent Or Entity?
- Step 3: Identify Entities
- Order Design Steps
- Score Your Own Bot Design
- Google Assistant Combines Many Unit 4 Ideas At Once
- Build The Conversation Flow
- Step 5: Errors & Bot Voice
- Bot Personality Styles
- Step 6: Test Your Bot
- Intent vs Entity
- Spot Language AI!
- Unit 4 Master Quiz
- Write Up The Bot You Designed
- Unit 4 Vocabulary
- Reflection Time
- Looking Ahead
- Can Language AI Do This Yet?
- Byte's Final Message
- Unit 4 Champion!
Requirements this topic helps meet
- ADEKChatbot mechanics, build a simple conversational agent
- CBSEDistinguish key predictive techniques - regression, classification, clustering - and explain key AI domains: Data Science, Computer Vision, Natural Language Processing; explain uses of AI in healthcare, education, transport, communication (Class 7).
- CISCE + statesDescribe the domain of Natural Language Processing (NLP) — text understanding, text generation, language translation, question answering/chatbots, dialogue systems — and internet search types (navigation, informational, transactional, investigative, voice)
- CISCE + statesApply Named Entity Recognition (NER) to classify names of people/organizations/locations in text, perform text summarization by extraction, and build a spam-email classifier using Machine Learning
- IBConstruct and trace algorithms expressed in pseudocode, independent of a specific programming language's syntax
- England + CambridgeDesign, use and evaluate computational abstractions that model the state and behaviour of real-world problems and physical systems
- England + CambridgeFollow and create algorithms presented as flowcharts using standard symbols (start, stop, process, decision, connector), including multiple variables
- England + CambridgeFollow, edit and create algorithms using flowcharts and pseudocode; use selection statements and Boolean logic; progress from linear to binary search
- CSTA (US)Use flowcharts and/or pseudocode to address complex problems as algorithms.
- AI4K12 (US)Progress from illustrating ambiguity via homophones/homonyms resolved by context (K-2), to illustrating pronoun reference ambiguity resolved by world knowledge (3-5), to illustrating prepositional-phrase attachment ambiguity resolved by meaning (6-8), to describing polysemy (multiple word senses) and how statistical NLP systems resolve it less reliably than human common sense (9-12).
- AI4K12 (US)Progress from demonstrating what an intelligent assistant like Siri or Alexa can and cannot do (K-2), to demonstrating question types a search engine or assistant can and cannot answer (3-5), to describing NLP tasks such as summarization, sentiment analysis, question answering, and machine translation (6-8), to describing NLP approaches ranging from keyword/template matching to deep neural networks, and when each is appropriate (9-12).
- OECD–EC (Europe)Be aware that some AI systems aim to provide human-like interaction (e.g. customer-service chatbots) and be able to interact with and give feedback to such a system to influence what it recommends next (AI examples 8 and 33, competence 2.1).
- SingaporeLearn with AI: apply AI to support and enhance one's own learning process.
- Australia + NZStudents and educators use generative AI tools in ways designed to enhance teaching and learning outcomes rather than replace core learning processes
Unit 5: AI Games and Puzzles 11 topics
AI Game Masters28 lessons· meets 3 requirements inEngland + CambridgeAI4K12 (US)OECD–EC (Europe)
Lessons
- The Carrom Championship
- Grandfather's Wisdom
- What Is Sam Doing In Those Extra Seconds?
- What Is Game AI Actually Doing?
- One Position Hides More Futures Than You Can Count
- What Kind of Game Is It?
- What Can the AI See?
- Why Does One Recipe Work On So Many Games?
- Three Steps, Three Traps
- Step 1: Read the Position
- Step 2: Imagine the Futures
- Step 3: Score and Choose
- Order the Loop
- Move Quality Lab
- The Chess App In Your Pocket Out-Calculates Everyone
- What Makes a Move Good?
- Rate the Moves
- Prune It or Keep Searching?
- Famous Game AI
- Match the Big Ideas
- Play Like an Engine
- Spot the Thinking Ahead
- Game AI Quiz
- Explain How An Engine Picks Its Move
- Game AI Vocabulary
- Key Takeaways
- Byte's Game Wisdom
- AI Game Master!
Requirements this topic helps meet
- England + CambridgeDesign, use and evaluate computational abstractions that model the state and behaviour of real-world problems and physical systems
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- OECD–EC (Europe)Deliberately assign parts of a task to human strengths versus AI capabilities (Manage 3, advanced).
Decision Trees Deep Dive31 lessons· meets 6 requirements inCSTA (US)AI4K12 (US)
Lessons
- The Perfect Dosa
- Mom's Decision Chart
- Chart, or Gut Feeling?
- How Does Mom's Chart Always Lead To The Right Step?
- What is a Decision Tree?
- Simple Questions Can Guide Surprisingly Smart Decisions
- Question Box or Action Box?
- Why Do Clear YES Or NO Questions Make Trees Stronger?
- Umbrella Decision Tree
- Build the Tic-Tac-Toe Tree
- Ludo AI Decision Tree
- Match the Tree Parts
- Good vs Bad Questions
- Sort Good vs Bad Questions
- 5-Step Decision Process
- Order the Steps
- Decision Tree Precision Lab
- Guided Apps Use Branching Questions Too
- Which Shot Does the Tree Play?
- Add a Question, or Prune It?
- Worth Taking?
- Learning from Mistakes
- Find the Bug!
- Spot Decision Trees!
- Decision Trees Quiz
- Explain How A Decision Tree Reaches An Action
- Decision Tree Vocabulary
- Key Takeaways
- Trees vs. Knowledge Graphs
- Byte's Design Test
- Decision Tree Master!
Requirements this topic helps meet
- CSTA (US)Understand that representation includes modalities (text, speech, audio, image, video) and symbolic mappings (text, graphs).
- CSTA (US)Create and evaluate an appropriate AI algorithm (e.g., a decision tree classifier) to accomplish a task.
- AI4K12 (US)Progress from giving examples of symbols encountered in daily life such as restroom icons and card suits (K-2), to explaining hierarchical tree structures such as ancestry trees and taxonomies (3-5), to translating a structure like a game board, road map, or mind map into a labeled graph of nodes and links (6-8), to translating the premises of a syllogism from English into predicate-logic notation and completing the inference (9-12).
- AI4K12 (US)Progress from drawing a tree by repeatedly splitting branches and placing data at nodes (K-2), to describing the parts of a tree (nodes, parent, leaves) and how they relate (3-5), to describing the parts of a labeled graph (nodes, links, source/target) and how trees are a special case of graphs (6-8), to describing how schemas structure information about people, places, or things within knowledge graphs used by search engines (9-12).
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from analyzing a decision-tree-building game and how the learning algorithm adds/replaces nodes (3-5), to comparing how decision-tree learning (add a node) differs from neural-network learning (adjust every weight) (6-8), to describing how decision-tree, neural-net, regression, and reinforcement-learning algorithms each adjust their own type of internal representation (9-12). K-2 is marked N/A.
If-Then-Else Logic29 lessons· meets 4 requirements inCBSEEngland + CambridgeAI4K12 (US)
Lessons
- The Ride-Share Algorithm
- The Driver's Logic
- Two Rules Are True At Once — Which One Wins?
- IF, THEN, ELSE — Which Action Runs?
- Tiny True Or False Checks Control Huge Systems
- Decision or Just a Step?
- Why Does One Clear Rule Feel So Powerful?
- Order the Tic-Tac-Toe Bot
- Run the Ludo Bot
- Can A Machine Answer It?
- Comparison Operators
- Match the Operators
- Logic Rule Lab
- Navigation Apps Ask Yes-Or-No Questions All The Way
- Does the Named Number Count?
- Combining with AND
- Combining with OR
- Using NOT
- Sort AND, OR, NOT
- Drive the Nested Cricket Bot
- Three Golden Rules
- Debug the Logic!
- Spot IF-THEN-ELSE!
- IF-THEN-ELSE Quiz
- Explain How IF-THEN-ELSE Makes A Choice
- IF-THEN-ELSE Vocabulary
- Key Takeaways
- Byte's Logic Wisdom
- Logic Master!
Requirements this topic helps meet
- CBSEInterpret and solve multi-step, multi-layered problems using advanced 3-D transformations, algebraic/multi-rule numerical patterns, interdependent-clue decomposition, and rule-based algorithms with conditional branching - increasing in complexity from Class 6 through Class 8 (e.g., Class 8 adds number-system generalisation across decimal/binary/ternary/Roman/Chinese numerals, and optimisation problems).
- England + CambridgeUnderstand simple Boolean logic (AND, OR, NOT) and its uses in circuits and programming; understand binary representation and carry out simple binary operations
- AI4K12 (US)Progress from giving examples of symbols encountered in daily life such as restroom icons and card suits (K-2), to explaining hierarchical tree structures such as ancestry trees and taxonomies (3-5), to translating a structure like a game board, road map, or mind map into a labeled graph of nodes and links (6-8), to translating the premises of a syllogism from English into predicate-logic notation and completing the inference (9-12).
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
The Chess Challenge34 lessons· meets 5 requirements inCSTA (US)AI4K12 (US)
Lessons
- The Chess Challenge
- Grandpa's Secret
- Why Did Grandpa Reject The Flashy Move?
- MAX, MIN And Your Score
- The Worst Case Is A Floor, Not A Forecast
- Keep It Or Throw It Away?
- Why Does Planning For The Worst Produce Great Moves?
- Think Simply
- Tic-Tac-Toe Example
- Step 1: All Moves
- Step 2: Predict
- Step 3: Best Move
- Minimax Depth Lab
- Stockfish Checks The Replies Before It Trusts A Move
- Order the Steps
- The Algorithm
- Work It Out
- Minimax Match
- Game Trees
- Speed-Up Tricks
- Inspect Your Own Minimax Tree
- Chess Scoring
- Sort Piece Values
- Connect Four AI
- Advantages
- Limitations
- Sort: Good vs Bad
- Beyond Minimax
- Spot the Minimax
- Quiz Time!
- Explain Why A Depth Limit Turns A Guarantee Into A Guess
- New Words
- Byte's Wisdom
- Strategy Champion!
Requirements this topic helps meet
- CSTA (US)Identify the assumptions inherent in the operation and output of an AI model and how these assumptions might have different implications for different people.
- CSTA (US)Describe the properties, biases, and assumptions of various kinds of AI models (e.g., classifier, predictor, recommender).
- AI4K12 (US)Progress from giving examples of symbols encountered in daily life such as restroom icons and card suits (K-2), to explaining hierarchical tree structures such as ancestry trees and taxonomies (3-5), to translating a structure like a game board, road map, or mind map into a labeled graph of nodes and links (6-8), to translating the premises of a syllogism from English into predicate-logic notation and completing the inference (9-12).
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from drawing a simple 3-node search tree showing two possible tic-tac-toe moves (K-2), to drawing a full search tree of next moves for tic-tac-toe or nim and picking the best move (3-5), to modeling breadth-first search on a graph problem such as common-friend detection (6-8), to illustrating and comparing breadth-first, depth-first, and best-first search algorithms for growing a search tree (9-12).
The School Tournament31 lessons· meets 10 requirements inUAEEngland + CambridgeAI4K12 (US)OECD–EC (Europe)+2
Lessons
- The School Tournament
- Humans Fight Back
- One Machine, Three Very Different Results
- The Three-Question Test
- The Same Game Can Swing Either Way
- Four Games AI Has Already Taken
- Why Written Rules Hand AI Such A Big Edge
- Five Machine Advantages
- Match: How Each Game Fell
- Name What Stops The Machine
- Six Human Edges
- Sort: Who Takes This One?
- The Three-Question Lab
- The Engine That Beat Everyone Now Reviews Your Games
- Close Battle: Hidden Cards
- Match: The AI They Slowed Down
- Close: Carrom, And The Problem Of Hands
- Sixty Years In Order
- What Moved, And What Hasn't
- Rule Of Thumb Check
- Five Honest Human Weaknesses
- Five Honest Machine Weaknesses
- Match: What Each Weakness Costs
- The Team-Up, In Order
- Three Real Team-Ups
- Spot the Winner
- Quiz Time!
- Explain Which Games AI Takes, And Which It Doesn't
- New Words
- Byte's Wisdom
- Game Analyst!
Requirements this topic helps meet
- UAECompare AI with humans (Cycle 2 part)
- England + CambridgeYoung people should understand how AI works, its capabilities and limitations, and learn to use it effectively without becoming dependent on it; this should be delivered primarily through Computing rather than assumed to develop automatically
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- OECD–EC (Europe)Break a problem into parts and consider AI support for specific steps (Manage 3, intermediate).
- OECD–EC (Europe)Describe an AI system's purpose, intended users, and basic constraints (Shape 1, intermediate).
- OECD–EC (Europe)States that a foundational understanding of AI's technical functioning and the ability to recognise chances, limits and risks are indispensable for students to use AI responsibly and reflectively ('digital mündig'); calls on the Länder to integrate AI-related informatics competences and 'Bildung in der digitalen Welt' competences into curricula for all subjects and school types, without itself specifying discrete grade-level student outcomes.
- SingaporeUnderstand AI's uses, risks and limitations, and apply ethical considerations (e.g. academic integrity, proper data handling) when using AI to support learning.
- SingaporeIdentify and explain AI's potential benefits, limitations and risks.
- SingaporeLearn beyond AI: develop distinctly human capabilities and dispositions that AI cannot replicate.
- Australia + NZBuild knowledge about the function and purpose of AI systems (Years 5-6), progressing to evaluating the advantages and disadvantages of AI and other emerging technologies (Years 7-8)
The Sudoku Champion29 lessons· meets 2 requirements inCBSEAI4K12 (US)
Lessons
- The Sudoku Champion
- Human vs AI Method
- How Does The App Test So Many Puzzle Paths So Fast?
- Puzzles vs Games
- A Hard Puzzle Becomes Easier When Rules Remove Possibilities
- 3 Puzzle Techniques
- Why Do Constraints Make Hard Puzzles Feel Smaller?
- Sudoku Rules
- Step 1: Analyze
- Step 2: Fill Obvious
- Step 3: Backtracking
- Order the Steps
- Puzzle Solver Lab
- Google Maps Solves Road Networks Like Search Problems
- Rubik's Cube AI
- 3 Cube Methods
- Why AI is Fast
- Maze Solving AI
- BFS Example
- Match Methods
- 5 General Strategies
- Sort Strategies
- Real AI Examples
- Spot the Strategy
- Quiz Time!
- Explain How Puzzle AI Narrows Down A Solution
- New Words
- Byte's Wisdom
- Puzzle Master!
Requirements this topic helps meet
- CBSEInterpret and solve multi-step, multi-layered problems using advanced 3-D transformations, algebraic/multi-rule numerical patterns, interdependent-clue decomposition, and rule-based algorithms with conditional branching - increasing in complexity from Class 6 through Class 8 (e.g., Class 8 adds number-system generalisation across decimal/binary/ternary/Roman/Chinese numerals, and optimisation problems).
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
The Delivery Challenge31 lessons· meets 7 requirements inCBSEEngland + CambridgeCSTA (US)AI4K12 (US)
Lessons
- The Delivery Challenge
- 3 Routes Instantly!
- Why Can A Navigation App Spot Strong Routes So Quickly?
- What Does 'Best' Mean?
- A Tiny Estimate Can Guide A Huge Map
- Where Pathfinding Hides
- Why Does A Good Heuristic Save So Much Searching?
- Method 1: Random Walk
- Method 2: Greedy Best-First
- Method 3: BFS
- Two Halves of the King
- A* Formula: F = G + H
- Pathfinding Route Lab
- Google Maps Recalculates Routes With Guided Search
- A* Decides: Open Or Skip?
- Order the Formula
- Heuristics Explained
- 3 Types of Heuristics
- Build and Compare a Search Maze
- Match Heuristics
- Walls, Costs and Rerouting
- 4 Advanced Techniques
- Real-World Challenges
- Pac-Man Ghost AI
- Sort Pathfinding Methods
- Spot the Pathfinding
- Quiz Time!
- Explain How Pathfinding Chooses A Smart Route
- New Words
- Byte's Wisdom
- Navigation Champion!
Requirements this topic helps meet
- CBSEInterpret and solve multi-step, multi-layered problems using advanced 3-D transformations, algebraic/multi-rule numerical patterns, interdependent-clue decomposition, and rule-based algorithms with conditional branching - increasing in complexity from Class 6 through Class 8 (e.g., Class 8 adds number-system generalisation across decimal/binary/ternary/Roman/Chinese numerals, and optimisation problems).
- England + CambridgeShow understanding of how graphs can be used to aid AI, and use A* and Dijkstra's algorithms to perform searches on a graph (candidates not required to write the search algorithms themselves)
- CSTA (US)Create and evaluate different abstract representations (e.g., subway map).
- AI4K12 (US)Progress from constructing a map of a home, school, or neighborhood and understanding that 'the map is not the territory' (K-2), to demonstrating how different map styles (geometric vs. topological) capture different aspects of the world (3-5), to showing how a game board (tic-tac-toe, chess) can be represented as a plain-language description of squares, relationships, and piece positions (6-8), to describing how to represent a concept as a schema.org-style schema with superconcepts and inherited properties (9-12).
- AI4K12 (US)Progress from giving examples of symbols encountered in daily life such as restroom icons and card suits (K-2), to explaining hierarchical tree structures such as ancestry trees and taxonomies (3-5), to translating a structure like a game board, road map, or mind map into a labeled graph of nodes and links (6-8), to translating the premises of a syllogism from English into predicate-logic notation and completing the inference (9-12).
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from drawing a simple 3-node search tree showing two possible tic-tac-toe moves (K-2), to drawing a full search tree of next moves for tic-tac-toe or nim and picking the best move (3-5), to modeling breadth-first search on a graph problem such as common-friend detection (6-8), to illustrating and comparing breadth-first, depth-first, and best-first search algorithms for growing a search tree (9-12).
The Cricket Strategy27 lessons· meets 5 requirements inCISCE + statesAI4K12 (US)Korea, Japan, HK
Lessons
- The Cricket Strategy
- Kabir's Smart Mix
- Why Does A Predictable Strategy Collapse So Fast?
- Random vs Deterministic
- A Little Randomness Can Protect A Strong Strategy
- When Random FAILS
- Why Does Pure Logic Become Easy To Counter?
- Exploration in Games
- Breaking Patterns
- Monte Carlo Methods
- 3 Mixing Strategies
- Exploration vs Exploitation
- Strategy Mix Lab
- YouTube Mixes Familiar Picks With Exploration
- Games Where Random Helps
- Games Where Smart Wins
- Match Random vs Smart
- The Ludo Rule Ladder
- Real World Examples
- Sort: Explore vs Exploit
- Spot the Strategy
- Order AI Learning
- Quiz Time!
- Explain Why Smart AI Still Needs Some Exploration
- New Words
- Byte's Wisdom
- Strategy Champion!
Requirements this topic helps meet
- CISCE + statesExplain the evolution of computing from deterministic (binary logic, conditional gates) to probabilistic approaches and why AI is needed for probabilistic, real-life problems
- CISCE + statesCompare automated vs autonomous systems and deterministic vs probabilistic decision making; contrast human decision making (subjective) with machine decision making (objective) via object-classification examples
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from modeling the use of a classification or search algorithm to solve a simple problem (K-2), to describing differences between decision-tree and neural-network classification algorithms (3-5), to comparing several algorithms that could solve a specific reasoning problem based on input data and decision complexity (6-8), to listing, for each of eight reasoning-problem types, an algorithm that could solve it (9-12).
- Korea, Japan, HKUse a simulation tool (e.g. a traffic-light decision simulator) to explore how an AI system makes sequential decisions under simulated conditions.
The School Game Fair25 lessons· meets 3 requirements inEngland + CambridgeAI4K12 (US)
Lessons
- The School Game Fair
- Rajiv's Design Steps
- Same Rules, Two Very Different Opponents
- Step 1: Rule or Design Choice?
- Rules Keep It Legal, Goals Give It A Character
- Step 2: Goals and Personality
- Why Do Tiny Behavior Tweaks Change How Smart AI Feels?
- Step 3: One Turn, Five Tools
- Step 4: Difficulty or Cheating?
- Example: Ghost AI
- Example: Card Game AI
- Example: Strategy Game AI
- Game AI Design Lab
- Pac-Man's Four Ghosts Are Four Different Personalities
- Step 5: Test Like a Player
- Common AI Mistakes
- Match AI Tools
- Order Design Steps
- Sort AI Personalities
- Spot the Good AI
- Quiz Time!
- Explain How You Would Design A Game Opponent
- New Words
- Byte's Wisdom
- Game Designer Champion!
Requirements this topic helps meet
- England + CambridgeDesign, use and evaluate computational abstractions that model the state and behaviour of real-world problems and physical systems
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from modeling the use of a classification or search algorithm to solve a simple problem (K-2), to describing differences between decision-tree and neural-network classification algorithms (3-5), to comparing several algorithms that could solve a specific reasoning problem based on input data and decision complexity (6-8), to listing, for each of eight reasoning-problem types, an algorithm that could solve it (9-12).
Unit 5 Review: AI Games & Puzzles24 lessons· meets 4 requirements inEngland + CambridgeAI4K12 (US)
Lessons
- Unit 5 Celebration!
- Chapters 41-43 Review
- Five Tools That Look Nothing Alike
- Chapters 44-46: Does It Take a Turn?
- Four Chapters, One Repeated Move
- Chapters 47-49: Build the Opponent in Order
- Strong Systems Stack These Ideas, They Do Not Choose Between Them
- The Six You Must Remember
- Match Chapters to Concepts
- True or False?
- Who Takes This Game?
- Order the A* Loop
- Build An Opponent From Three Dials
- A Robot Vacuum Runs Three Of These Chapters At Once
- Match the Methods
- Calculate It, or Mix In Randomness?
- Ultimate Quiz Part 1
- Ultimate Quiz Part 2
- Explain Which Unit 5 Tools Your Machine Would Need
- Outside Games: Match the Idea
- Vocabulary Review
- Achievement Checklist
- Byte's Final Words
- Unit 5 Complete!
Requirements this topic helps meet
- England + CambridgeUnderstand that robots can work autonomously and describe robotics benefits in manufacturing/food production
- AI4K12 (US)Progress from illustrating a next possible board state in tic-tac-toe (K-2), to illustrating how a computer represents game play as a linear sequence of board positions connected by legal moves (3-5), to illustrating how a computer solves a maze, finds a map route, or reasons over a knowledge graph by drawing a search tree (6-8), to identifying real-world problems (task planning, scheduling, resource allocation) as search problems and describing their states and operators (9-12).
- AI4K12 (US)Progress from identifying a problem as a classification problem or a search problem (K-2), to categorizing problems as classification vs. search using examples like Teachable Machine (3-5), to categorizing problems as classification, prediction, combinatorial search, or sequential decision problems (6-8), to categorizing real-world problems across eight reasoning-problem types including heuristic search, adversarial search, logical deduction, and statistical inference (9-12).
- AI4K12 (US)Progress from modeling the use of a classification or search algorithm to solve a simple problem (K-2), to describing differences between decision-tree and neural-network classification algorithms (3-5), to comparing several algorithms that could solve a specific reasoning problem based on input data and decision complexity (6-8), to listing, for each of eight reasoning-problem types, an algorithm that could solve it (9-12).
How Robots Sense, Decide and Move28 lessons· meets 16 requirements inADEKCISCE + statesEngland + CambridgeSingapore
Lessons
- The Library Rover
- Sense, Decide, Act: Level Up
- Robot, Machine or Body-less AI?
- Is Byte a Robot?
- Two Circles: Robots and AI
- What Did the AI Add?
- How a Robot Senses
- A Sensor for Each Job
- How a Robot Moves
- Bend the Elbow
- Draw the Rover's Loop
- Program a Robot Reflex
- Which Rule Wins?
- How Much Does the Robot Decide?
- Level Up the Car
- Who Is in Control?
- Robotaxis in Their Service Area
- Safe by Design
- Stop in Time
- Safe or Not?
- Robots in Six Places
- Advantage or Drawback?
- Case: The Rover Bumps a Student
- A Robot Helper at Home?
- Checkpoint: Robots
- Design Your Rover
- Robot Words
- Robot Engineer!
Requirements this topic helps meet
- ADEKRobotics / physical device
- CISCE + statesExplain what a robot is, its characteristics, and the benefits of robots over humans (quality, productivity, efficiency, working span, hazardous environments)
- CISCE + statesState a brief history of robotics evolution since 1900 and recite Isaac Asimov's Three Laws of Robotics
- CISCE + statesClassify robots by terrain/field (aerial, ground, underwater), by control (manual, automatic), and identify bio-inspired robots (humanoid, bird, snake, insect)
- CISCE + statesGive at least one real-world application example of robots in domestic, industry, medical, defence, entertainment and agriculture domains
- CISCE + statesDescribe the general block diagram and building blocks of a robot (mechanical, electronic, computational) and identify robots by demonstration/video (industrial arm, humanoid, mobile)
- CISCE + statesDistinguish types of motion (linear, angular, circular) and types of joints/links (prismatic, revolute, spherical; rigid, soft) and use links and joints to create specific motion
- CISCE + statesDefine degrees of freedom of a robot and identify them through illustration
- CISCE + statesExplain New Age Robotic Systems (warehouse, assistant, smart-home/school/mobility robots, autonomous cars/drones, medical/healthcare robotics) and why NARS are becoming increasingly relevant and possible
- CISCE + statesDifferentiate a machine from a robot, and a robot from a cobot (human-in-the-loop, multi-robot systems), tracing the progress from robots to cobots
- CISCE + statesExplain gears (role, principles, gear ratio) and classify sensors (internal: position/velocity/acceleration/force; external: contact/non-contact) and their functions in a robotic system
- CISCE + statesDescribe actuators (linear, rotary), control systems (manual vs automatic) and integrate sensors, actuators and controller in a robotic system using a block diagram (input-controller-robot-feedback)
- England + CambridgeUnderstand that robots can work autonomously and describe robotics benefits in manufacturing/food production
- England + CambridgeDescribe how sensors, microprocessors and actuators combine to create automated systems, and describe advantages/disadvantages for given scenarios (industry, transport, agriculture, weather, gaming, lighting, science)
- England + CambridgeUnderstand what is meant by robotics, describe the characteristics of a robot (mechanical structure, electrical components/sensors/actuators/microprocessors, programmability), and the roles robots perform (industry, transport, agriculture, medicine, domestic, entertainment) with advantages/disadvantages
- SingaporeUnderstand what a smart/intelligent robot is and how AI enables it to respond to inputs.
Unit 6: AI Project Lab 10 topics
Aisha's Big Idea33 lessons· meets 26 requirements inUAEADEKCBSECISCE + states+7
Lessons
- Aisha's Big Idea
- Ms. Sharma's Lesson
- The Missing Piece
- The 5 P's, In Order
- A Tiny Plan, A Bigger Dream
- P1: PURPOSE
- Strong vs Weak Goals
- Why Planners Feel Calmer
- P2: PLAN
- Milestone or Just a Task?
- Order the Weeks
- Planning Balance Lab
- Trello Turns Plans Into Cards
- P3: PREPARE
- Sort the Resources
- P4: PERFORM
- Order Project Tasks
- P5: PRESENT
- Match the 5 P's
- Two Ways Plans Die
- Fix All Five Traps
- In Scope, Out of Scope
- What the Three Plans Share
- Scope Check Challenge
- Byte's 5 Tips
- Match the Tips
- Real AI Developers
- Planning Quiz
- Explain the 5 P's
- Vocabulary
- Ready to Start?
- What's Next?
- Planning Master!
Requirements this topic helps meet
- UAEDesign AI projects for real-world problems
- ADEKBuild an AI solution to a real problem
- CBSEDescribe the stages of the AI project cycle (Define Problem, Collect Data, Test AI Tools, Reflect and Improve); apply no-code tools to tackle real-world problems and reflect on their utility/effectiveness; analyse contributions of AI to fields like healthcare, automation, and education, understanding both benefits and risks; describe AI ethics as values/guidelines for responsible AI (Class 8).
- CBSEDefine Statistical Data and its applications; define No-Code and Low-Code AI and identify differences from coded AI; use a no-code tool (Orange Data Mining) to perform data exploration, modeling, and evaluation, relating stages to the AI project cycle (Unit 4: Statistical Data, assessed via practicals only).
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- IBUse and apply technology effectively to access, process, and communicate information, and to model and create solutions to problems
- CSTA (US)Decompose problems and subproblems into parts to facilitate the design, implementation, and review of programs.
- CSTA (US)Distribute tasks and maintain a project timeline when collaboratively developing computational artifacts.
- CSTA (US)Collaborate with many contributors through strategies such as crowdsourcing or surveys when creating a computational artifact.
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Be interested in experimenting with various types of AI systems depending on personal needs (e.g. virtual assistants, image analysis, speech/face recognition, robots) (AI example 71, competence 5.2).
- UNESCOPractise skills to identify and scope a problem that could be solved by building or training a new AI model, based on a simulation project (CG4.1.4.2).
- UNESCODevelop skills to assess an AI system's needs for data, algorithms, programming languages, software, and computing hardware, including feasibility given regulatory/ethical restrictions and cost (CG4.1.4.3).
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- SingaporeKeep digital knowledge current, including awareness of emerging technologies such as AI, as one of the nine core NDLP digital competencies.
- SingaporeDevelop digital information management skills as one of the nine core NDLP digital competencies.
- SingaporeDevelop digital communication, collaboration and participation skills as one of the nine core NDLP digital competencies.
- SingaporeApply the Find-Think-Apply-Create framework when locating, evaluating, using and creating with digital information (exact breakdown of the four components not published on the primary MOE page reviewed).
- SingaporeDevelop digital literacy and technological skills needed to thrive in an increasingly technology-driven world, as a central goal of the Masterplan.
- Australia + NZApply digital literacy skills (investigating with, creating with, communicating and collaborating with, and managing/operating digital tools) across all learning areas, including recognising the role of AI-enabled tools in everyday digital systems
The Great Project Choice40 lessons· meets 25 requirements inUAEADEKCISCE + statesIB+8
Lessons
- The Great Project Choice
- Three Student Stories
- The Four-Week Trap
- The 3 Golden Rules
- A Smaller Project Can Win Bigger
- Rule 1: DOABLE
- Easy vs Hard Data
- Why Small Categories Help
- Rule 2: INTERESTING
- Interest Categories
- Rule 3: MEANINGFUL
- Sort by Impact
- How Countries Choose AI Problems
- Productivity or Wellbeing?
- Strategy Check
- Project Fit Meter
- Google Lens: One Job Per Mode
- Match the 3 Rules
- Category 1: Images
- Category 2: Sound
- Category 3: Pose
- Category 4: Creative
- Match Project Types
- 5-Step Selection Process
- Order Selection Steps
- 5 Common Mistakes
- Three Choices That Worked
- Feasibility Check
- Byte's Idea Filters
- Match the Filters
- Scoping in the Real World
- Your Own Problem Instead
- Ready or Not Yet?
- Project Selection Quiz
- Explain Your Project Choice
- Vocabulary
- Decision Checklist
- What's Next?
- How Your Project Is Judged
- Project Chooser!
Requirements this topic helps meet
- UAEAI applied across sectors
- UAEDesign AI projects for real-world problems
- ADEKBuild an AI solution to a real problem
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- IBGiven a design situation, identify a problem that needs to be solved and analyse the need for a solution
- IBConduct an inquiry into the nature of a design problem to inform the design brief
- IBWrite a detailed design specification that drives the development of a solution
- IBDevelop and present a proposed design solution derived from the specification
- CSTA (US)Use sensors to collect data, and then train an AI model using the sensor data.
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Propose specific design changes to improve an AI system for human well-being (Shape 4, intermediate).
- OECD–EC (Europe)Be interested in experimenting with various types of AI systems depending on personal needs (e.g. virtual assistants, image analysis, speech/face recognition, robots) (AI example 71, competence 5.2).
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- SingaporeApply computational thinking and design-thinking concepts to create a solution for a real-world problem (secondary school students).
- SingaporeTrain a robot or AI-enabled prototype to respond to external signals through hands-on exploration and tinkering.
- Korea, Japan, HKConstruct and apply an AI system using a variety of data ([9정04-03]).
- Korea, Japan, HKDesign, develop and present a group AI project (e.g. an AI-robotics solution to an everyday/pandemic-related problem such as an automatic soap dispenser or light switch) applying knowledge, skill and — in the culminating Booklet-3 project — attitudes from the whole module.
- ChinaStudents apply a 'needs analysis → technology matching → outcome evaluation' engineering-thinking framework to AI projects
- Australia + NZStudents understand that generative AI tools should be used in ways that benefit all members of the school community and support wellbeing, not just individual convenience
Kabir's Excitement34 lessons· meets 22 requirements inUAEADEKCISCE + statesCSTA (US)+4
Lessons
- Kabir's Excitement
- Diagnose Meera's Disaster
- Why Did Meera's AI Fail?
- The Research Solution
- Research Saves Build Time
- Now, or After Testing?
- Problem Meets Rule
- The Hidden Power of Clean Data
- Type 1: Topic Research
- Type 2: Technical Research
- Type 3: Similar Projects
- What Each Question Saves You
- Emma's Bird Identifier
- Rohan's Three Kinds of Research
- Research Readiness Lab
- Google Photos Needs Organized Images Too
- 5 Common Mistakes
- Sort Research Habits
- Rebuild Meera's Sweet Sorter
- Sort the Photos
- Do These Two Overlap?
- Why Document?
- Research Log Template
- Order Research Steps
- Ready to Shoot?
- Byte's 5 Tips
- How Professionals Do It
- Spot Good Research
- Research Quiz
- Explain Research First
- Research Vocabulary
- Key Takeaways
- What's Next?
- Chapter Complete!
Requirements this topic helps meet
- UAELearn from data; algorithms make decisions
- UAEEthics, cultural values and inclusion guide AI decisions
- ADEKTrain/test a model, report performance
- ADEKExamine dataset gaps/bias
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CISCE + statesDistinguish supervised, unsupervised and reinforcement learning, and build a machine-learning model (dataset preparation, training, testing) such as one recognizing Indian Sign Language (ISL) gestures
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- CSTA (US)Describe the ways that bias can be introduced and mitigated in an AI model.
- CSTA (US)Using a dataset and a machine learning pipeline, develop an AI model, and consider the impact of the model on various users.
- CSTA (US)Explore strategies to turn ethical considerations into actions, such as mitigating bias in datasets.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from illustrating why training a classifier for a broad concept like 'dog' requires large, diverse image data by browsing datasets such as ImageNet or Coco (3-5), to illustrating how objects in images are segmented and labeled with bounding boxes to build an object-recognition training set (6-8), to evaluating a real AI training dataset's size, acquisition/labeling method, storage cost, and time-to-produce, including where bias can enter (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from examining a labeled dataset to identify problems that could cause incorrect predictions, e.g. an all-green 'healthy foods' set (K-2), to examining training-data features/labels for potential sources of bias such as Amazon's biased resume sorter (3-5), to explaining how unbalanced training data biases a classifier, e.g. faces or loan applications (6-8), to investigating demographic imbalances in training data using data-visualization tools such as histograms or Pandas (9-12).
- AI4K12 (US)Progress from creating a model card for a classification or prediction model documenting its intended use, training data, accuracy, and limitations (3-5), to constructing an ethical matrix of stakeholders and their values for an AI system and evaluating alignment (6-8), to designing an AI system using a full ethical design process that considers all stakeholders' values across fairness, transparency, explainability, accountability, privacy, and societal values (9-12). K-2 is marked N/A.
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Explain how biased data or design choices create unfair outcomes (Engage 6, intermediate).
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- Australia + NZConsider the possible bias or harm to others that may occur when inputting data into AI systems, and explore how AI systems use predictive strategies based on large amounts of data to perform tasks or solve problems
Lily's Big Moment34 lessons· meets 19 requirements inADEKCISCE + statesIBCSTA (US)+4
Lessons
- Lily's Big Moment
- Day 1: Make Lily's Calls
- The Hidden Shortcut
- Days 2-3: Training
- Smart, or Panic?
- Days 4-5: Success!
- Data Collection Checklist
- Plan the Photo Session
- 4 Data Problems
- Match Problems & Fixes
- Better Examples Beat Fancy Tricks
- Organizing Your Data
- Teachable Machine Setup
- Order Setup Steps
- How Training Works
- Prototype Accuracy Lab
- Snapchat Lenses Need Strong Training Too
- Train Your Own Tiny Model
- Testing Your AI
- Sort Confidence Levels
- Honest Test or Flattering Test?
- Match Symptom & Fix
- The Improvement Cycle
- 4 Quick Improvements
- Export Without Losing It
- Byte's 6 Tips
- Real-World Building
- Spot Good Practice
- Building Quiz
- Explain the Prototype Mindset
- Building Vocabulary
- Key Takeaways
- What's Next?
- Chapter Complete!
Requirements this topic helps meet
- ADEKTrain/test a model, report performance
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CISCE + statesBuild a program (using block-based tools such as PictoBlox) that recognizes facial expressions using a pre-trained model, and explore gesture-based software applications for accessibility
- CISCE + statesDistinguish supervised, unsupervised and reinforcement learning, and build a machine-learning model (dataset preparation, training, testing) such as one recognizing Indian Sign Language (ISL) gestures
- IBPlan the creation of the chosen solution and follow that plan to build a prototype sufficient for testing and evaluation
- CSTA (US)Use sensors to collect data, and then train an AI model using the sensor data.
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- CSTA (US)Using a dataset and a machine learning pipeline, develop an AI model, and consider the impact of the model on various users.
- CSTA (US)Create a program using available AI tools, AI plugins, APIs, and/or AI models, with ethical considerations for the model's end users and others impacted: fairness, bias, and accuracy, and then create a model card.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- SingaporeTrain a robot or AI-enabled prototype to respond to external signals through hands-on exploration and tinkering.
- SingaporeDesign and build a prototype that incorporates artificial intelligence to address a problem.
- Australia + NZStudents understand that their privacy and personal data must be protected when using generative AI tools, and are taught not to input sensitive personal or others' data into such tools without appropriate safeguards
The 70% Problem35 lessons· meets 16 requirements inADEKCISCE + statesIBCSTA (US)+4
Lessons
- The 70% Problem
- Day 1: Classmate Testing
- The Hidden 70% Problem
- Feedback Gold
- Improvements Round 1
- Why Failure Is the Most Useful Result
- Day 3: Family Testing
- Family Insights
- The 91% Success!
- Why Testing Matters
- 4 Types of Testing
- Match Testing Types
- Users See What Builders Miss
- 5 User Test Steps
- User Test Steps
- Feedback Questions
- Pattern Analysis
- 4 Common Patterns
- Match Problems & Solutions
- The Feedback Loop
- 4 Improvement Strategies
- Feedback Loop Lab
- Spam Filters Learn From Corrections
- Edge Cases
- Trust the Reading?
- Pattern Detective
- Realistic Goals
- Week 3 Schedule
- Byte's 6 Tips
- Testing Checklist
- Real-World Connection
- Checkpoint Quiz
- Explain Why Testers Matter
- New Vocabulary
- Testing Master!
Requirements this topic helps meet
- ADEKTrain/test a model, report performance
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CISCE + statesDistinguish supervised, unsupervised and reinforcement learning, and build a machine-learning model (dataset preparation, training, testing) such as one recognizing Indian Sign Language (ISL) gestures
- IBPlan the creation of the chosen solution and follow that plan to build a prototype sufficient for testing and evaluation
- CSTA (US)Refine computational models based on the data they have generated.
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- CSTA (US)Using a dataset and a machine learning pipeline, develop an AI model, and consider the impact of the model on various users.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- OECD–EC (Europe)Assess an AI system's performance using defined criteria and benchmark tests (Shape 2, intermediate).
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- SingaporeDesign and build a prototype that incorporates artificial intelligence to address a problem.
Two Presentations34 lessons· meets 19 requirements inCISCE + statesIBEngland + CambridgeCSTA (US)+4
Lessons
- Two Presentations
- Sofia's Presentation
- Why Did One Class Lean In?
- The Difference
- A Hook Can Change the Whole Room
- The Shape of a Talk
- Match the Parts
- Why Stories Beat Lists
- Part 1: The Hook
- Sort the Hooks
- Part 2: The Content
- Order the Content
- Audience Attention Lab
- TED Speakers Rehearse the Flow
- Part 3: The Closing
- Poster Tips
- Slide Tips
- Demo Preparation
- 3-Step Demo Process
- Handling Problems
- Beating Nervousness
- Voice Tips
- Body Language
- Types of Questions
- Answer Formula
- Rehearsal Strategy
- Practice Tips
- Presentation Checklist
- Byte's Last Tips
- Real-World Connection
- Checkpoint Quiz
- Explain a Strong Presentation
- New Vocabulary
- Presentation Pro!
Requirements this topic helps meet
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- IBUse and apply technology effectively to access, process, and communicate information, and to model and create solutions to problems
- IBWhere a school offers digital design (rather than or alongside product design), demonstrate digital-tool competencies such as prototyping, graphic/web design, and multimedia editing through the same four-criterion design cycle
- England + CambridgeUndertake creative projects selecting, using and combining multiple applications across devices to achieve challenging goals, including collecting and analysing data and meeting the needs of known users
- England + CambridgeCreate, re-use, revise and re-purpose digital artefacts for a given audience, with attention to trustworthiness, design and usability
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- SingaporeKeep digital knowledge current, including awareness of emerging technologies such as AI, as one of the nine core NDLP digital competencies.
- SingaporeDevelop digital information management skills as one of the nine core NDLP digital competencies.
- SingaporeDevelop digital communication, collaboration and participation skills as one of the nine core NDLP digital competencies.
- SingaporeApply the Find-Think-Apply-Create framework when locating, evaluating, using and creating with digital information (exact breakdown of the four components not published on the primary MOE page reviewed).
- SingaporeDevelop digital literacy and technological skills needed to thrive in an increasingly technology-driven world, as a central goal of the Masterplan.
- Australia + NZApply digital literacy skills (investigating with, creating with, communicating and collaborating with, and managing/operating digital tools) across all learning areas, including recognising the role of AI-enabled tools in everyday digital systems
Your Building Week!34 lessons· meets 16 requirements inADEKCISCE + statesCSTA (US)AI4K12 (US)+2
Lessons
- Your Building Week!
- Week 1 Goals
- How Can Five Days Build an AI?
- Monday: Data Day
- Small Daily Wins Build Big Projects
- Tuesday: Finish Photos
- Wednesday: Build Day!
- Thursday: Improve It
- Friday: Polish Day
- Match Days to Tasks
- Why Timers Beat Endless Studying
- 25-Minute Focus
- Focus Steps Order
- Week Planner Lab
- Notion Keeps a Build Log
- Workspace Setup
- Sort Workspace Tips
- Asking for Help
- Daily Progress Tracker
- Five Problems, Five Fixes
- Smart Move or Slip-Up?
- Stay Inspired
- Real Help or Crossing the Line?
- Teacher Checkpoints
- Weekend Work (Optional)
- End of Week Reflection
- Green Light or Fix It?
- Byte's Week 1 Wisdom
- What's Next?
- Week 1 Knowledge Check
- Explain Week 1 Workflow
- Key Vocabulary
- How Your Project Is Judged
- Week 1 Complete!
Requirements this topic helps meet
- ADEKTrain/test a model, report performance
- ADEKBuild an AI solution to a real problem
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CISCE + statesBuild a program (using block-based tools such as PictoBlox) that recognizes facial expressions using a pre-trained model, and explore gesture-based software applications for accessibility
- CISCE + statesDistinguish supervised, unsupervised and reinforcement learning, and build a machine-learning model (dataset preparation, training, testing) such as one recognizing Indian Sign Language (ISL) gestures
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- CSTA (US)Using a dataset and a machine learning pipeline, develop an AI model, and consider the impact of the model on various users.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- Korea, Japan, HKConstruct and apply an AI system using a variety of data ([9정04-03]).
- Korea, Japan, HKDesign, develop and present a group AI project (e.g. an AI-robotics solution to an everyday/pandemic-related problem such as an automatic soap dispenser or light switch) applying knowledge, skill and — in the culminating Booklet-3 project — attitudes from the whole module.
Refinement Week!29 lessons· meets 23 requirements inADEKCISCE + statesIBEngland + Cambridge+5
Lessons
- Refinement Week!
- Week 2 Goals
- Why Did One Project Pull Ahead?
- Monday: Measure It
- Tiny Fixes Stack Into a Whole Project
- Tuesday: Improve!
- The Refinement Loop
- Fresh Eyes Find Hidden Problems
- Fix the Cause, Not the Number
- Fair Test or Fooling Yourself?
- Refinement Lab
- Why Polish Is a Real Skill
- Poster and Demo Script
- Poster Rescue
- Order Your Demo
- What to Rehearse
- Being Nervous
- Calm Toolkit
- Emergency Calm
- Night Before
- Buddy Reports
- Feedback Exchange
- Week 2 Rescue
- Week 2 Check
- Week 2 Quiz
- Explain Refinement
- Key Vocabulary
- Byte's Wisdom
- Week 2 Complete!
Requirements this topic helps meet
- ADEKTrain/test a model, report performance
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CISCE + statesBuild a program (using block-based tools such as PictoBlox) that recognizes facial expressions using a pre-trained model, and explore gesture-based software applications for accessibility
- CISCE + statesDistinguish supervised, unsupervised and reinforcement learning, and build a machine-learning model (dataset preparation, training, testing) such as one recognizing Indian Sign Language (ISL) gestures
- IBDesign tests to evaluate a solution, carry out those tests, and objectively evaluate the solution's success
- IBIdentify areas where a design solution could be improved and explain the solution's impact on the client or target audience
- IBWhere a school offers digital design (rather than or alongside product design), demonstrate digital-tool competencies such as prototyping, graphic/web design, and multimedia editing through the same four-criterion design cycle
- England + CambridgeUndertake creative projects selecting, using and combining multiple applications across devices to achieve challenging goals, including collecting and analysing data and meeting the needs of known users
- England + CambridgeCreate, re-use, revise and re-purpose digital artefacts for a given audience, with attention to trustworthiness, design and usability
- CSTA (US)Seek and incorporate feedback from team members and users to refine a solution that meets user needs.
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- CSTA (US)Describe the ways that bias can be introduced and mitigated in an AI model.
- CSTA (US)Using a dataset and a machine learning pipeline, develop an AI model, and consider the impact of the model on various users.
- CSTA (US)Explore strategies to turn ethical considerations into actions, such as mitigating bias in datasets.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- Korea, Japan, HKUse a simulation tool (e.g. a traffic-light decision simulator) to explore how an AI system makes sequential decisions under simulated conditions.
- Australia + NZEvaluate existing and student-developed digital solutions and information systems, and their suitability for personal, environmental and social criteria, against needs and requirements
The Big Day!29 lessons· meets 12 requirements inADEKCISCE + statesCSTA (US)AI4K12 (US)+2
Lessons
- The Big Day!
- Setup Time
- Two Demos Freeze. One Room Stays Impressed.
- Presentation Flow
- The Glitch Is Not the Problem
- Aisha's Bird AI
- Project Variety
- The Room Changes the Presenter
- Maya's Turn
- Kabir's Energy!
- Showcase Confidence Lab
- How YouTube's Caption AI Plans for Being Wrong
- Handle Problems
- Great Audience
- Ask Good Questions
- Star and Wish
- Receive Feedback
- Celebrate You!
- Showcase Awards
- Real Prizes
- Save Memories
- Reflect
- Gratitude
- Showcase Check
- Showcase Quiz
- Write a Star and a Wish
- Key Vocabulary
- Byte's Message
- Showcase Complete!
Requirements this topic helps meet
- ADEKBuild an AI solution to a real problem
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
- Korea, Japan, HKConstruct and apply an AI system using a variety of data ([9정04-03]).
- Korea, Japan, HKDesign, develop and present a group AI project (e.g. an AI-robotics solution to an everyday/pandemic-related problem such as an automatic soap dispenser or light switch) applying knowledge, skill and — in the culminating Booklet-3 project — attitudes from the whole module.
Full Circle!6 lessons· meets 9 requirements inCISCE + statesCSTA (US)AI4K12 (US)UNESCO
Lessons
- Full Circle!
- Six Units, One Year
- Which Job Is It?
- Name the Pattern
- Final Quiz!
- GRADE 7 COMPLETE!
Requirements this topic helps meet
- CISCE + statesImplement a simple image classifier distinguishing categories of objects (e.g. cats vs dogs, fruits vs vegetables) and build a facial-recognition application that can associate identified faces with known individuals
- CISCE + statesTrack the movement of objects across a video sequence over time (e.g. a ball in sports footage, vehicles in traffic surveillance)
- CSTA (US)Using sensor data (e.g., from autonomous vehicles), train an AI model.
- AI4K12 (US)Progress from demonstrating, with instructor help, how to train a computer to recognize something using a tool like Teachable Machine (K-2), to training a classification model and examining its accuracy on new inputs (3-5), to training and evaluating a classification or prediction model on a tabular dataset (6-8), to using supervised or unsupervised learning on real-world data and quantitatively evaluating the results on a nontrivial test set (9-12).
- AI4K12 (US)Progress from creating a labeled dataset with explicit features to classify things like foods or toys (K-2), to creating a labeled dataset with multiple feature types (discrete, binary, continuous) and training a classifier with a tool such as Machine Learning for Kids (3-5), to creating a decision-tree training dataset and exploring how different feature encodings affect the resulting tree (6-8), to comparing two real-world datasets in terms of their features and encodings and the consequences for learning (9-12).
- AI4K12 (US)Progress from describing and using AI extensions/plugins in a familiar programming framework such as Scratch, e.g. speech-to-text, face recognition (3-5), to creating a novel application using AI extensions/plugins in a chosen framework such as Scratch, MIT App Inventor, or Calypso (6-8), to creating a novel application using AI tools of the student's choice, now commonplace and freely available without advanced degrees or expensive equipment (9-12). K-2 is marked N/A.
- UNESCOCritically evaluate and leverage free/open-source AI datasets (e.g. MNIST, CIFAR, ImageNet) and AI/algorithm libraries (e.g. Teachable Machine, PyTorch, Keras) to address authentic, task-based problems (CG4.2.3.3).
- UNESCOTransfer values, knowledge and skills to craft a task-based AI tool by customizing an existing AI model or toolkit, including modifying/creating a dataset for real-world contexts (e.g. scraping data with tools like BeautifulSoup, cleaning/encoding/preprocessing it) (CG4.3.3.1).
- UNESCOEnhance creativity in customizing AI toolkits and coding — leveraging AI development platforms, enhancing datasets, and modifying programming code (including open-source options) to design and test creative task-based AI tools (CG4.3.3.2).
Bring the course to your school or region
Schools, school groups, resellers and partners: we will walk you through the alignment for your board and the lessons behind it.
Our own reading of our own course, ratings reviewed 22 October 2026. It is not an endorsement, certification or accreditation by any authority named here, and it makes no claim about learning results.