Context

Artificial intelligence (AI) is becoming an increasingly visible part of everyday life. From recommendation systems and digital assistants to translation tools and image generators, AI technologies are transforming how we communicate, learn, and solve problems.

Youth encounter AI daily through digital tools such as personalized social media feeds, video and music suggestions, photo filters, search autocomplete, and video game features. As these technologies are already part of youth’s everyday experiences, instructors can help them build awareness and reflect on how AI works, even in STEM workshops, camps, or programs that are not specifically focused on AI. AI can also raise important questions and concerns for youth. Some may feel excited about its possibilities, while others may worry about its impact on society. You play a key role in guiding reflection and fostering the confidence needed to navigate AI thoughtfully. 

As AI becomes more present in education and society, you are also engaging with it in different ways. You may be coming into this training with your own experiences, questions, or viewpoints about AI, whether that includes excitement, uncertainty, or even hesitation about its role. This module is designed to support you in navigating these perspectives while building your understanding of AI in educational content. 

AI literacy is an important part of preparing youth for the future. Rather than just learning to use AI tools, it involves gaining the knowledge, skills, and mindset to understand and interact with AI in meaningful ways. By exploring how AI works, recognizing its limitations, evaluating its benefits and risks, and using it responsibly , youth can ask critical questions about technology, understand its societal impact, and recognize that humans ultimately design and guide these systems.

Developing AI literacy is not important for youth, but also for you as an instructor. Building your own understanding of AI can support you in navigating these perspectives, making informed decisions about its use, and facilitate meaningful experiences for youth.

Activity 7: AI Awareness

To begin exploring AI literacy, test your knowledge about how Canadians are thinking and talking about AI.

This short quiz is based on findings from Actua’s “Ready for AI?” report, which surveyed over 1,000 Canadian students, 500 educators, and 1,000 parents and caregivers to better understand how people across the country are experiencing and thinking about AI. 

Quiz Link: actua.ca/aiready/quiz 

After completing the quiz, take a few minutes to reflect on the following questions:

• What surprised you about how youth are using or thinking about AI?
• How might the results influence the way you introduce or discuss AI with youth?
• What role can you play in helping youth move beyond simply using AI tools to thinking more critically about them?

What is Artificial Intelligence?

Artificial Intelligence refers to machines and computer systems designed by people that use data and algorithms to identify patterns, make predictions, and generate outputs such as recommendations, responses, or decisions. These systems can perform tasks such as recognizing images, recommending content, translating language, or generating text and visuals.

Although AI has gained significant public attention in recent years, the concept has existed for decades. Researchers have long explored how computers could recognize patterns, make predictions, and learn from data. Today’s AI tools are part of this longer evolution. Developing a foundational understanding of AI helps us better understand how these systems work, how they are applied in the world, and how they can be used thoughtfully and responsibly.

Review the resource below to explore examples of AI, understand what makes a machine artificially intelligent, and clarify common misconceptions.

Open these slides in a new tab: Understanding AI.

Activity 8: AI-Mazing Machines

Explore the following stations by completing the steps provided for each tool. After exploring each station, reflect on the questions below. You may use your Activity Workbook to record your thoughts.

• How might this tool help youth better understand what AI is and what it is not?
• How could this be incorporated in different programming types (camps, clubs, workshops) and for different ages of youth?

Station One

Tool: Google - AutoDraw

Type of AI: Machine Learning

Steps: 

1. Launch Google AutoDraw, a drawing tool that uses AI to recognize sketches and suggest polished images.
2. Draw a simple object using the AutoDraw tool (for example: a cat, tree, or bicycle)
3. Observe the suggestions that appear based on your sketch.
4. Select one of the suggested images to replace your drawing.
5. Try drawing several different objects and notice how the suggestions change.

Observe: Did the tool always recognize what you were trying to draw? When it didn’t, what do you think caused the mismatch?

Station Two

Tool: Google - TestFx

Type of AI: Generative AI

Steps: 

1. Launch Google TextFX, an AI tool that generates new words and phrases from prompts.
2. Enter a word or short phrase (for example: ocean, technology, or climate).
3. Select one of the available tools (such as similes, rhymes, or word associations) to generate new text.
4. Explore different tools and experiment with new words or phrases. 
5. Try modifying your input and observe how the generated results change.

Observe: In what ways might a tool like this support creativity or brainstorming? Are there moments where the results feel repetitive, unexpected, or less meaningful?

Station Three

Tool: Google - Say What You See

Type of AI: Generative AI

Steps: 

1. Launch Google Say What You See, a game that demonstrates how AI systems interpret written prompts to generate images.
2. Observe the image displayed on the screen.
3. Write a caption that describes the image as clearly as possible.
4. Submit your caption and observe the image generated by the AI.
5. Compare the generated image with the original and notice how your description influenced the result.

Observe: How closely does the generated image match the original image? What kinds of problems might arise if an AI system misinterprets a prompt or description?

How does AI Learn? 

Most modern AI systems learn through a process called machine learning, where algorithms improve their performance by analyzing data.

Instead of being programmed with exact instructions for every situation, a machine learning system is trained using examples. By analyzing patterns in these examples, the system learns relationships in the data, allowing it to make predictions about new information it has not come across before.

For example:

• An image recognition system may be trained using thousands of labelled pictures of cats and dogs.
• A recommendation system may analyze which songs users listen to most frequently.
• A translation system may learn patterns from millions of sentences written in multiple languages.

Data is the foundation of this process. Data can include numbers, images, text, audio, and other types of information. Large collections of data, known as datasets, are used to train AI systems so that algorithms can identify patterns and produce outputs such as predictions, classifications, or generated responses. Because AI systems learn from data, the quality and diversity of the data used can influence how well the system performs and the kinds of results it produces.

AI systems also rely on human-designed rules and criteria that determine how information is analyzed. AI developers decide which features or characteristics are important and how much weight they should have in the system’s decisions. These design choices can influence the outcomes produced by the algorithm and may unintentionally introduce assumptions or bias.

Understanding how AI learns from data can help support conversations about where training data comes from and how it shapes the results AI systems produce. AI systems are often trained using large datasets collected from online sources, images, text, and other digital information created by people. This means the data used to train AI can reflect human behaviours, perspectives, and biases. Discussing these ideas with youth can help them think more critically about their digital footprint, how information they create online may be used to train AI systems, and why the quality and diversity of data matter when developing and using AI technologies.

Review the resource below to explore how machine learning works and examine algorithmic bias and approaches to address it. 

Open these slides in a new tab: Data and Machine Learning.

Activity 9: Inside Algorithmic Decisions

Explore the Most Likely Machine (mostlikelymachine.artefactgroup.com), which demonstrates how algorithms make decisions and how bias can emerge from the choices people make when designing them.

1. Launch the Most Likely Machine.
2. Choose which historical figures you think would most likely win specific awards (Example: Go to a Top University, Go Viral, or Biggest Troublemaker).
3. Identify traits that might influence these outcomes (Example: adaptability, playful, funny).
4. Rank these traits in order of importance so the algorithm can prioritize them.
5. Run the algorithm and review which historical figures the system predicts are most likely to win each award.
6. Compare the algorithm’s predictions with your original choices. 

Reflection Questions

Take a few moments to reflect on the following questions:

• How did the traits and rankings you selected influence the algorithm’s results? What assumptions or biases might have shaped those choices?
• How might algorithms like this be used within larger AI systems that make decisions in areas such as social media, hiring, or recommendations?
• How might you encourage youth to question how algorithms influence the information, recommendations, or content they view online?

How Does AI Create?

Recent advances in AI have led to tools that can generate new content, including text, images, music, and video. These are often referred to as generative AI systems.

Generative AI systems are trained on large collections of existing content. By analyzing patterns in this data, the system learns how elements such as words, images, sounds, or styles are commonly arranged. Once trained, the system can use these learned patterns to produce new outputs based on a prompt or input provided by a user

Although these systems can produce impressive results, they do not truly understand the content they generate. Instead, they predict patterns that are statistically likely based on the data they were trained on.

For example:

• Text generators analyze patterns in language to produce written responses.
  • Examples:Google Gemini and Microsoft Copilot
• Image and video generators learn visual patterns to create new visual media based on prompts.
  • Examples: Google Veo and Canva Magic Media
• Music generators learn rhythms and melodies to compose new audio.
  • Examples: Suno and ElevenLabs

Generative AI relies on large datasets to learn these patterns. These datasets may include books, images, recordings, or other digital content created by people. The outputs produced by generative AI are not copied directly from these examples but are generated based on the patterns the system has learned.

Many generative AI tools respond to prompts, which are instructions or questions provided by a user. The quality and clarity of a prompt can influence the results produced by the system. When youth experiment with generative AI tools, encourage them to ask specific questions, provide clear instructions, and try different prompts to compare results. This helps demonstrate that human input plays an important role in shaping AI-generated outputs.

As AI-generated media becomes more common, it is important to help youth question and evaluate the content they encounter. For example, youth can consider whether the content includes unusual patterns or inconsistencies, whether sources or references are provided, and whether the information can be confirmed using other reliable sources.

Understanding how generative AI creates content can help youth think more critically about the tools they use and recognize both the possibilities and limitations of AI-generated media.

Review the resource below to explore how generative AI works and how to identify AI-generated media. 

Open these slides in a new tab: Generative AI and AI-Generated Media.

Activity 10: Two Truths and AI

Explore Two Truths & AI (commonsense.org/games/two-truths-and-ai), an interactive game that asks you to identify which images were generated by AI. 

1. Launch Two Truths & AI
2. Review the three movie posters displayed on the screen. Two posters are real and one was generated by AI.
3. Examine each poster closely and choose which one you think is AI-generated. Search for clues such as unusual details, strange text, or visual inconsistencies.

Take a few moments to reflect on the following questions:

• What details made the AI-generated image easier or harder to identify?
• How might AI-generated images influence what people believe or share online?
• How could you encourage youth to question, verify, and think critically about the images or media they encounter online?