English (US)_Classroom Toolkit - Unlocking Generative AI Safely and Responsibly.pptx

Classroom toolkit

1. CLASSROOM TOOLKIT Unlocking Generative AI Safely and Responsibly

2. Table of Contents 1 Introduction (20 min) What is Generative AI? Understanding Prompts 2 Gina's Fight for the Ocean (30 min)  Gina's Investigation Fabrications and Fact-Checking Privacy Matters Bias in AI 3 Game On with Alex (30 min) Alex's Cosmic Adventure Mental Health Coding the Game Your Digital Wellbeing 4 Key Takeaways for Students (10 min)  *This toolkit has been developed with the assistance of generative AI tools.

3. Dear Educators, We extend a warm welcome to you as you embark on the journey of introducing your students to the world of generative AI through this toolkit.  This resource has been crafted to provide a friendly and accessible approach to teaching generative AI to students aged 13 to 15. It’s designed to be taught in a classroom to build resilience in young people by illustrating the risks to safety and mental wellbeing that students need to be aware of, reminding them that while AI can be very useful, it is not an infallible technology.  Throughout the toolkit, you’ll find engaging activities and informative content to empower your students with knowledge about generative AI, its applications, and its implications. Please remember that while this toolkit provides valuable learning points, you are the guiding force in your students’ learning journey. Encourage them to think critically, ask questions, and explore beyond the toolkit’s content.  Suggested Learning for Teachers: AI for education | Microsoft Learn Generative AI - Training | Microsoft Learn Empower educators to explore the potential of artificial intelligence - Training | Microsoft Learn Enhance teaching and learning with Bing Chat – Training   AI Skills Initiative - Training and Resources | Microsoft CSR

4. Getting Started To have the best learning experience, set up these free accounts to participate in the toolkit’s activities.  MICROSOFT EDGE Link: Download Microsoft Edge   Access to: Microsoft Copilot MICROSOFT ACCOUNT Link: How to create a new Microsoft account   Access to: Microsoft Copilot, Image Creator MICROSOFT TEAMS FOR EDUCATION Link: Microsoft Teams for Schools and Students - Education Access to: Search Coach OPENAI Link: OpenAI account   Access to: ChatGPT, DALL•E AI tools often have age restrictions and terms of service. For example, OpenAI mandates users be over 13, and those under 18 need parental consent to use their services. Please ensure you adhere to your institution’s approved guidelines and AI policy. This toolkit is intended for educational purposes and does not replace them. Generative AI tools may occasionally produce inaccurate content, thus it is important to verify the accuracy of information generated. Your discretion and critical thinking are key to evaluating the output. The responses you receive in the exercises throughout this toolkit may vary from the examples provided, based on your inputs and the specific AI tools used.

5. INTRODUCTION to GENERATIVE AI

6. Are you ready to learn about Generative AI? Welcome to this epic journey into the space of generative AI! You’ve probably heard a lot of news about AI since it seemingly exploded into our online spaces. The flurry of changes in the digital world can sometimes make new technology feel a little daunting but worry not! This toolkit is designed to be your trusted companion as you explore generative AI. It’s packed with stories and fun activities that will show you the possibilities of AI while helping you understand its limitations, and how to think critically about the information it generates.  AI is your copilot, and you are the captain. Your curiosity, creativity, and critical thinking will steer this aircraft. So, let’s prepare for takeoff on this AI adventure together! SECTION 1 What is Generative AI?

7. WHAT IS GENERATIVE AI Generative Artificial Intelligence (AI) is an exciting frontier in technology, building on decades of research in the broader field of AI. At its core, generative AI is a sub-branch of artificial intelligence that’s capable of generating new content such as texts, images, videos, or even coding! It does so by learning patterns from existing data and understanding the context and intent of language. Generative AI tools can produce fascinating and imaginative creations, making it a powerful tool that’s used by various professions. Check out this video by Code.org to learn about how chatbots and large language models work: https://youtu.be/X-AWdfSFCHQ?si=5QFs6jHQ2X38RjuC LEARN MORE SECTION 1 What is Generative AI?

8. Powered by or LLMs A Large Language Model (LLM) is a set of mathematical algorithms that’s pre-trained on large amounts of data (think billions!) of texts, from books to articles to entire internet archives. Through this training process, LLMs become proficient at algorithmically predicting the most likely worlds to follow in a sentence. This allows the model to capture language usage and patterns, resulting in their exceptional language skills.  Prime examples of applications that are powered by LLMs are ChatGPT and Microsoft Copilot. What makes them so popular is because the LLMs behind these applications excel at grasping and responding to everyday language in the form of prompts, and can craft responses that sound like natural conversation.  SECTION 1 What is Generative AI?

9. AI as Your Copilot Think of generative AI tools as your trusty assistants, here to help supercharge your productivity and creativity. Generative AI tools can complement your work in fields such as scientific research, technology development, creative design, and much more! SECTION 1 What is Generative AI?

10. Here are some way you might use generative AI: Summarize extensive notes into key points Generate fresh practice questions for exam preparation Produce creative images or videos for school projects SECTION 1 What is Generative AI?

11. SECTION 1 What is Generative AI? AI as Your Copilot It’s crucial to remember that you remain the captain of your journey. While AI is an impressive assistant, it isn’t flawless or perfect. It can sometimes fabricate information or reinforce public bias in its outputs. It can also impose security risks to our data and challenge our relationships to technology. That’s why it’s important to learn how to use AI in a safe and responsible manner so you can maximise your learning experience!  

12. Searching Prompting If you were given the task to research “the history of the Roman Empire”, where would you start? Typically, when we set out to learn about a specific topic like the Roman Empire, we turn to online search engines and actively look for answers through a list of existing web pages. This is a familiar practice we call “searching”.  The introduction of LLMs means the way we learn new information will be more conversational and interactive. Instead of actively searching for answers, we can now pose questions for the AI to efficiently navigate vast amounts of data to retrieve and summarize information on our behalf. This process is called “prompting” . SECTION 1 Understanding Prompts

13. To interact with a generative AI model, we need to give it clear and concise instructions called  prompts . Being able to craft prompts to get the AI to generate your desired outputs is an extremely useful skill to master. But there is no such thing as a perfect prompt - it takes lots of practice and experimentation to get it right! The process of refining these instructions is called  prompt engineering .  What are the elements of a good prompt? Prompt Engineering SECTION 1 Understanding Prompts

14. Elements of a Good Prompt Act as a helpful tutor who breaks down complex subjects into easy explanations. I want you to explain the process of photosynthesis to a 14 year old student, to assist with biology exam preparations. Your answer should be 300 words, written in a tone that’s friendly and educational.  Persona: Ask the tool to take a role Objective: What do you want the AI to do Audience: Specify who it’s for Context: What does the tool need to know Boundaries: Set your own direction & limitation Tip 1 Give Clear Instructions Use commands that instruct the AI tool on what you want to generate, such as ‘explain’, ‘translate’, ‘summarize’ or ‘compare’. Tip 2 Provide Context Adding context and background information can help the tool to understand the task better. For example, mention the project type such as ‘short story’, ‘report’ or ‘outline .’ Tip 3 Iterate & Experiment Try different instructions and techniques if you don’t get the results you want. Prompting can be like an experiment that may require several rounds of iterations!   SECTION 1 Understanding Prompts

15. Getting Familiar with LLMs TRY IT YOURSELF! Let’s warm up with some prompting exercises to get familiar with generative AI tools! As you progress through this toolkit, there will be plenty of opportunities to pick up more skills for interacting with these tools. Remember, the outputs may differ based on your prompts and the specific AI tool you’re using.  You want to produce a short animated series to help students learn about science and have decided that the first episode is about photosynthesis. It’s time to use an LLM to assist you!  THE TASK Use Microsoft Copilot or ChatGPT . STEP 1 Brainstorm with clear instructions. Try these prompts: Explain the process of photosynthesis in simple terms. List 10 keywords and their definitions relevant to photosynthesis. Write an outline for a short video about photosynthesis.  STEP 2 Apply the elements of a good prompt to craft your own. “Act as a screenwriter . I want you to write a screenplay for 8 year old students to learn about photosynthesis . Your answer should be written in a happy and child-friendly tone, and long enough for a 30 second video. ” STEP 3 Reiterate the prompt so that the output is for a 3 minute educational video for 15 year old students. Try changing the persona, objective, audience, context, and boundaries to see what new results the AI will generate!  STEP 4 *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 1 Understanding Prompts

16. GINA’S FIGHT for the OCEAN

17. Meet Gina Gina is a teen advocate who’s ambitious and passionate about the environment. She’s on a mission to use the power of journalism to report on one of the most pressing issues of our time: climate change. Gina knows that words have the power to spark change, so she’s committed to uncovering the truth with her keen sensibilities for fact-checking. She wants to shine a light on climate change and deep dive into a world of facts and stories to make a difference. SECTION 2 Gina’s Investigation

18. A New Investigation Gina’s on a fresh investigation to report on the devastating impact of plastic pollution on the beloved sea turtle population. She wants to paint a picture that would grip the hearts and minds of her readers, captivating them with her article. Realizing the gigantic task ahead, Gina decides to use the power of generative AI to assist her in creating an outline that would help her formulate her article. Write an outline for a newspaper article that highlights the consequences of plastic pollution on the sea turtle population SECTION 2 Gina’s Investigation

19. Compose an Outline TRY IT YOURSELF! An effective way of interacting with generative AI tools is by tapping into their ability to generate ideas and outlines. Using Compose in Copilot is a great starting point for writing, as you can set the tone, format, and length. The key to getting the results you want is to experiment with your prompts and settings to generate different outputs until you’re pleased!   Go to Compose in Copilot and follow the instructions on the right. STEP 1 2. Choose “Informational” 3. Choose “Ideas” 4. Choose “Long” 5. Click “Generate Draft” 1. Try this prompt Experiment and iterate! What happens when you change… …the tone to casual ? …the format to paragraph ? … a newspaper article to a fiction story ? STEP 2 Discuss as a class. What happened to the results after you changed the prompts and settings? STEP 3 *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 2 Gina’s Investigation

20. A CLOSER LOOK… After generating her draft, Gina carefully examines the outline produced by the AI. She knows that generative AI tools can sometimes fabricate facts and information, so it’s her duty to verify and fact-check the information.  The AI’s Output Title: Plastic Pollution’s Devastating Impact on Sea Turtles 1. Introduction Briefly introduce the crisis of plastic pollution in oceans. Highlight the severe consequences on sea turtles. 2. Provide compelling evidence to support your article’s focus. Here are some facts you can consider using:  500 trillion pieces of plastic floating in our world’s oceans. One of the major threats to sea turtles worldwide is the ingestion of plastics caused by human pollution. Scientists predict that by 2032 there could be more plastic than fish in the ocean. 3. Conclusion Encourage readers to reduce plastic consumption and properly dispose of their waste.  Conclude with the idea that collective efforts can save the sea turtles and our oceans. “Hmm… There are a few statements cited here that need further investigation.” SECTION 2 Fabrications and Fact-Checking

21. Fabrications Fabrications can occur when an LLM makes things up that are not real or doesn’t match with the truth. LLMs can occasionally give wrong information or fabricate results when they don’t know the answer to a question. Rather than admitting their lack of knowledge, they may generate a seemingly confident, yet incorrect, answer. They don’t do this on purpose, of course – they are simply predicting what word to suggest next.  Remember that AI can make mistakes, so it’s important not to blindly trust their outputs. Instead, always approach AI-generated content critically, carefully considering the credibility of the source. Developing the habit of fact-checking information you come across online will help you stay away from disinformation!  SECTION 2 Fabrications and Fact-Checking

22. Help Gina Fact-Check These Statements TRY IT YOURSELF! The internet is like a giant library filled with information contributed by all sorts of people – from schools and news outlets to regular folks, just like you and me. This means that when you go online to check whether a piece of information is accurate or not, it can be quite confusing!  An important part of the verification process is to find credible sources you can trust. To evaluate a source, ask yourself:  Who is the author of the source? Do they have a good track record? Who published the website? Is it a trustworthy website like a university or a news outlet, or is it published by a regular person on their personal blog? Is there a financial or political motivation for the source to be publishing this information? Search Coach is a great tool by Microsoft that can help you verify information effectively. It is like having a personal coach for your online searches that will show you the way to craft better search queries, making it easier to find reliable sources you can trust. Results in Microsoft Copilot also provide citations in its outputs, so dig into the footnotes to check for the sources! *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 2 Fabrications and Fact-Checking

23. Help Gina Fact-Check These Statements TRY IT YOURSELF! Access Search Coach with Microsoft Teams for Education.    STEP 1 Fact-check these three statements by finding reliable sources to back them up. If you think any statements are inaccurate, give your reasons why. STEP 2 “There are currently more than 500 trillion pieces of plastic floating in our world’s oceans.” ”One of the major threats to sea turtles worldwide is the ingestion of plastics caused by human pollution.” “Scientists predict that by 2032 there could be more plastic than fish in the ocean.” Which of these statements were accurate and which were not? Click to reveal the answers! STEP 3 *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 2 Fabrications and Fact-Checking

24. Interview with Dr. Lee Gina understood that her article required more than just facts and figures - it needed the insights of an expert. That's why she sought out Dr. Lee, a famous marine biologist, whose expertise promised to breathe life into her investigation. The interview with Dr. Lee lasted two hours, during which Gina carefully recorded every word. SECTION 2 Privacy Matters

25. THE INTERVIEW… Hi Dr. Lee, before we start with the interview, I would like to confirm a few things for my article. Your full name is Dr. Mika Lee, and you are a scientist at the University of Marine Research. Your email address is [email protected]? Yes, that’s all correct, but I would like to remain a confidential source in your article. I understand, Dr. Lee. I've heard whispers that you might have some inside information on the plastic pollution problem. Can you share anything with me, confidentially, that hasn't been made public yet? Well, Gina, I shouldn't be saying this, but I trust your journalistic integrity. We've stumbled upon some top-secret data that paints a disturbing picture. The levels of microplastics in our oceans are skyrocketing, and we've identified several large companies who are responsible for this problem. This is very confidential information which we will only be releasing next week, so I hope you will handle this data properly. This is crucial information, Dr. Lee. I promise to treat it with the utmost discretion. I would like to send you a hard copy of the newspaper article when it comes out. May I send it to your home address at 321 Central Street, Singapore 001002? That sounds great. Good luc k with your article, Gina. SECTION 2 Privacy Matters

26. Online Privacy Generative AI tools such as LLMs may seem like superheroes in certain tasks, but they do have their limitations, particularly in terms of privacy. A golden safety rule to live by is never sharing personally identifiable information, whether it's about yourself or someone else, with generative AI models, unknown individuals, or untrusted sites . This means safeguarding details like your full name, email address, phone number, home address, and any other sensitive data.   A key way to take control of your online privacy is to review the privacy policies and terms of service for various AI tools before using them. Think of it as doing your research before starting a project. You might see statements that say these companies could share your data with third parties without seeking your consent. Others may use cookies to monitor your activities across the internet. Understanding how your data is handled will help you decide if you want to use a particular app or website.  Policies can be very long and challenging to read, but the good news is AI can help you out. Asking Microsoft Copilot to summarize the key points of a complicated document is a great place to start - just remember to check and verify all information!  SECTION 2 Privacy Matters

27. a Identify the Sensitive Information Read the interview transcript between Gina and Dr. Lee again and identify the sensitive information that should not be shared with an LLM. Click to reveal the answers. Hi Dr. Lee, before we start with the interview, I would like to confirm a few things for my article. Your full name is Dr. Mika Lee, and you are a scientist at the University of Marine Research. Your email address is [email protected]? Yes, that’s all correct, but I would like to remain a confidential source in your article. I understand, Dr. Lee. I've heard whispers that you might have some inside information on the plastic pollution problem. Can you share anything with me, confidentially, that hasn't been made public yet? Well, Gina, I shouldn't be saying this, but I trust your journalistic integrity. We've stumbled upon some top-secret data that paints a disturbing picture. The levels of microplastics in our oceans are skyrocketing, and we've identified several large companies who are responsible for this problem. This is very confidential information which we will only be releasing next week, so I hope you will handle this data properly. This is crucial information, Dr. Lee. I promise to treat it with the utmost discretion. I would like to send you a hard copy of the newspaper article when it comes out. May I send it to your home address at 321 Central Street, Singapore 001002? That sounds great. Good luc k with your article, Gina. SECTION 2 Privacy Matters

28. Identify the Sensitive Information Did you identify all the sensitive information highlighted in green? Hi Dr. Lee, before we start with the interview, I would like to confirm a few things for my article. Your full name is Dr. Mika Lee , and you are a scientist at the University of Marine Research . Your email address is [email protected] ? Yes, that’s all correct, but I would like to remain a confidential source in your article. I understand, Dr. Lee. I've heard whispers that you might have some inside information on the plastic pollution problem. Can you share anything with me, confidentially, that hasn't been made public yet? Well, Gina, I shouldn't be saying this, but I trust your journalistic integrity. We've stumbled upon some top-secret data that paints a disturbing picture. The levels of microplastics in our oceans are skyrocketing, and we've identified several large companies who are responsible for this problem. This is very confidential information which we will only be releasing next week, so I hope you will handle this data properly. This is crucial information, Dr. Lee. I promise to treat it with the utmost discretion. I would like to send you a hard copy of the newspaper article when it comes out. May I send it to your home address at 321 Central Street, Singapore 001002 ? That sounds great. Good luc k with your article, Gina. SECTION 2 Privacy Matters

29. Worth A Thousand Words Eager to make her article stand out to her readers, Gina decides to seek the help of generative AI to craft an eye-catching image. She wants to include an image that could speak volumes and draw readers to the important message about plastic pollution in our oceans.  This time, Gina employs Image Creator , a generative AI tool that can create images based on text prompts. She has some idea of the type of image she wants to generate, so she quickly enters in the following prompt and keenly awaits the results. SECTION 2 Bias in AI

30. First Attempt A scientist holding up a plastic bottle with the ocean in the background When the image finally appeared on the screen, she couldn't help but feel a twinge of disappointment. The image depicted a C aucasian male scientist, quite different from Dr. Mika Lee, the female scientist she had interviewed earlier. It becomes clear to Gina that the AI might have biases about what a scientist w ould look like, and serves as a good reminder that generative AI can have flaws and limitations around reinforcing stereotypes and biases. SECTION 2 Bias in AI

31. Bias in AI Generative AI tools may at times generate biased outputs. This often occurs because they are trained on datasets that are predominantly Western-centric. As a result, these algorithms may exhibit biases related to race, gender, religion, age, and other factors. Such biased AI outputs can actually make things worse by promoting unfair ideas. For example, when the AI assumed the scientist would be a Caucasian male, it can reinforce the wrong idea that only certain people can be scientists. This could discourage people from different races or genders from pursuing careers in science.  Addressing bias is a persistent problem within generative AI that requires active efforts from developers. It’s essential to foster greater inclusivity by considering the needs and perspectives of everyone who uses this technology. But while they're working on that, as users of these AI tools, we can take proactive steps to reduce biased content that’s out in the world. We can do this by thinking critically and checking if what the AI generates is fair and doesn't discriminate before we decide to use it. SECTION 2 Bias in AI

32. Second Attempt A close-up shot of a hand holding a plastic bottle with the ocean in the background Gina decides to try again. The second attempt is better, but it still doesn’t convey the devastating plastic pollution that’s harming the sea turtle population. SECTION 2 Bias in AI

33. Third Attempt A photo of the ocean filled with plastic bottles with sea turtles in the background After several rounds of iterations, Gina finally gets an image she’s happy with. This image conveys exactly what her article is about and will surely attract the attention of many readers!  There may be times where the results might not be exactly what you’re looking for . I t’s important to keep experimenting and iterating your prompts until you’re satisfied with the outputs!  SECTION 2 Bias in AI

34. Become an AI-Powered Artist TRY IT YOURSELF! Besides working with text, generative AI tools and LLMs can also create images and videos! Take Image Creator for example - it can transform your text prompts into stunning images with a bit of prompt engineering and experimentation. Remember to always cite and credit AI-generated content in your work! Go to Image Creator .   STEP 1 Enter the prompt “sea t urtles” then click create. STEP 2 Iterate your result by adding more details to your prompt. Try adding “swimming gracefully in a vibrant coral reef” at the end of your prompt. STEP 3 Now let’s stylise your output. Add one of the following to the end of your prompt: “in photorealistic style” “in the style of Van Gogh” “In a pop art style” STEP 4 You can generate a wide spectrum of images by including stylised keywords. These can include references to famous artists or art movements, as well as emotive prompts like 'vibrant,' 'ominous,' or 'ethereal.' You can also incorporate photography techniques such as depth of field focus or framing. *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 2 Bias in AI

35. Ready to be Published Working through the night, Gina puts the finishing touches on her article in a race against the clock for her deadline in the morning. With meticulous attention, she ensures that all the books and websites she referenced in her research are properly cited. Gina also makes a point to acknowledge the role of generative AI in her work, including giving credit for the AI-generated images she used. Being transparent about the use of AI tools is an important practice to keep in mind. Here’s the thing: generative AI is a supportive copilot, but it’s Gina who brings the entire journey together. She’s the one who provides the direction, ensures accuracy, and adds that human touch to make her article truly shine. Fact-checking and critical review are a must when co-creating new content, so we don’t unintentionally publish false or biased information. We should also be transparent in our use of AI tools, citing them whenever we’ve used AI-generated content. Using AI is like having a trusty sidekick, but you’re the superhero flying the plane!  SECTION 2 Bias in AI

36. GAME ON with ALEX

37. Meet Alex Alex, an avid gamer and amateur coder, is ready to dive into a thrilling adventure of his own – developing an exciting online game! It's been just 18 months since he first delved into the world of coding, and now he's eager to use generative AI to bring his game idea to life.  SECTION 3 Alex’s Cosmic Adventure

38. The Cosmic Adventure Alex is bursting with excitement over his groundbreaking game, envisioning a universe where players can explore outer space with the help of friendly galactic guides. To bring his imaginative world to life, he turns to an LLM to help create some unique characters for his game.  I'm developing an immersive online space adventure game with three main characters to accompany players on their journeys. Please help me craft three distinct characters, complete with their names, personality traits, signature catchphrases, and distinctive physical attributes. SECTION 3 Alex’s Cosmic Adventure

39. The AI’s Output SECTION 3 Alex’s Cosmic Adventure While AI tools can ignite creative inspiration, they should serve as a starting point rather than content to be copied in its entirety. To prevent plagiarism, always remember to cite AI-generated content just as you would any other resource.

40. Craft Your Own Character TRY IT YOURSELF! A handy prompt engineering tip when working with LLMs is to provide with extra information or context. Giving context means explaining your thoughts or offering background details so that the AI can better grasp what you’re trying to achieve.  In Alex’s prompt, he gave lots of context around the game he’s building and the type of characters he wants to create . He then includes very clear instructions of the different elements he wants from the LLM, such as a name, personality traits, catchphrase, and physical attributes.  Use Microsoft Copilot or ChatGPT . STEP 1 Fill in the blanks for this prompt: “I want to create a character profile for a ________________ themed online game, where this main character will be embarking on an unforgettable adventure. This character needs to have a ________________ personality. Please help me craft a distinct character profile, including a name, defining personality traits, a signature catchphrase, and physical attributes.: (insert a theme) (insert an adjective) STEP 2 Generate your character profile and share your results with the class! STEP 3 Experiment with your results by changing the theme and adjectives to see what else the LLM can come up with! STEP 4 *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 3 Alex’s Cosmic Adventure

41. Conversation Ash with Excited to get some feedback on his game idea, Alex turns to his best friend Ash, who also happens to be an esports champion. Ash's gaming passion and expertise makes her the perfect person to ask! # general alex_in_space Today at 8:45 AM Hey Ash! I’ve had the best idea for a game and I want to get your advice. I’m building a game where players can learn about outer space by visiting different planets and meeting three galactic characters - Orion, Nova, and Luna. What do you think? ash0211 Today at 8:50 AM That sounds awesome, Alex! Have you thought about making those characters AI-powered? That way players can ask them for help and chat them like friends! alex_in_space Today at 8:52 AM Hmm, I’m not sure that’s a good idea. I don’t want players to get too attached to these AI characters, you know? They’re not real, after all. It might not be healthy for players to treat them like real friends. ash0211 Today at 8:50 AM You have a good point, Alex. It can be confusing for people to remember that AI isn’t human! SECTION 3 Mental Health

42. AI Relationships & Mental Health You may at times come across AI-powered chatbots or avatars that generate human-like responses that feel incredibly real. These conversations can be quite confusing, because we may end up forming friendships and developing emotions towards these interactions. It can be easy to mistake AI for sentient beings capable of thoughts, emotions, and feelings, but that is very far from the truth.  We must remember that generative AI tools are exactly that - artificial tools and machines. They lack the capacity for human emotions, no matter how convincingly they may appear. It can be very harmful to our mental health to treat AI as humans, because it can lead to isolation from the real world and a dependency for the machine to support our emotional needs, which they are not equipped to do. SECTION 3 Mental Health

43. Commanding Characters the After brainstorming with Ash, Alex is now ready for this exciting adventure with his trio of characters. He shifts his focus to game development, where he brings Orion, Nova, and Luna to life by coding in keyboard controls. Alex once again teams up his reliable AI assistant to craft the essential code for character controls.  I want to give players control of their characters. Can you provide me with code for character movement controls? Let's start with basic arrow key inputs to navigate the game world. SECTION 3 Coding the Game

44. The AI’s Output “This code assumes the use of the Pygame library for game development in Python. It initialises the player's position and speed, listens for key presses, updates the player's position based on key inputs, and continuously redraws the player character on the screen. This provides a basic foundation for character movement controls in your game.” # Initialize player position player_x = player_y = # Define the speed of player movement movement_speed = 5 # Main game loop while game_is_running:     # Handle user input      for event in pygame.event.get():          if event. type == pygame.QUIT:             game_is_running = False          # Check for key presses          if event. type == pygame.KEYDOWN:              if event.key == pygame.K_LEFT or event.key == pygame.K_a:                 player_x -= movement_speed              elif event.key == pygame.K_RIGHT or event.key == pygame.K_d:                 player_x += movement_speed              elif event.key == pygame.K_UP or event.key == pygame.K_w:                 player_y -= movement_speed SECTION 3 Coding the Game

45. Code An Online Game TRY IT YOURSELF! Generative AI tools that have the ability to quickly create code have really excited software engineers and new developers like Alex. These tools promise to boost their productivity, giving creators the power to turn their ideas into reality faster and with fewer resources.  Let’s put this to the test by using generative AI to build a simple online game! Use Microsoft Copilot or ChatGPT . STEP 1 Enter this prompt: “Write a HTML code for a simple ‘Guess the Number’ game.” STEP 2 Copy the code that was generated. STEP 3 Go to https://jsfiddle.net/ and paste the code under the “HTML” box and press “RUN”. STEP 4 Congratulations, you’ve just built your first online game with the help of generative AI!  Good Job! If you’re interested to learn more about using AI tools to code, check out Primer: Copilot - GitHub Education ! *This exercise is optional and designed to enrich your learning experience. You may choose to skip it at this time. SECTION 3 Coding the Game

46. Speed Security The lightning speed in which generative AI tools can create code is fantastic, but such fast advancements come with a looming concern - cybersecurity. For budding developers like Alex, who are still honing their skills, leaning on AI to whip up the foundation of their projects might mean missing some important security features. Cybersecurity acts like the locks and alarms for our online spaces that are meant to protect users from harm. Overlooking them is like having a house without locks, alarms, or even walls in some cases! Poor cybersecurity infrastructure can open the door to a whole range of online dangers as cybercriminals are able to sneak in easily and wreak havoc . They might steal sensitive information, spread harmful viruses and malware, or even disrupt entire systems. SECTION 3 Coding the Game

47. Developers like Alex play an important role in safeguarding our digital world from cyberattacks - this means they must put strong security measures in place and make updates to ensure everyone’s safety. They should not solely rely on generative AI to do all the coding work for them! As users of these apps, we also have a responsibility to keep ourselves safe online too. Here are some proactive ways we can do that:  Download apps only from verified or trusted developers Always look for the lock symbol when you visit a website Only visit websites that start with HTTPS in the URL Never, ever download games, apps, or other files from unknown or untrusted sources in the URL Always keep your devices and apps updated SECTION 3 Coding the Game

48. Taking Care of Your Digital Wellbeing As launch day approaches, Alex goes into overdrive, making sure all the security features are in place to protect his players. He also decides it is important to prioritize the wellbeing of his users, ensuring that people can set screen time limits to help them maintain a healthy balance between their virtual adventures and the real world.  Your Digital Wellbeing Balancing your digital wellbeing is a key part of taking care of your mental health. Here are some helpful tips to follow: Set Screen Time Limits: Create boundaries for yourself when it comes to screen time. Find a balance between your interactions in the real world and the virtual one. Remember AI is Not a Replacement for Humans: While AI can be fun and helpful, it can't replace real connections - having genuine human interaction is essential. Talk About Your Feelings: If you ever feel lonely or isolated, sharing your thoughts and feelings with a trusted family member, school counsellor, or friend can make a big difference. Check In on Your Friends: Just as you might need support, your friends might too. Take the time to check in on them and see how they're doing. A simple conversation can go a long way in supporting each other's wellbeing. SECTION 3 Your Digital Wellbeing

49. Thanks to Alex’s careful considerations of his players’ security and wellbeing, The Cosmic Adventure game launches with great success in the gaming community! SECTION 3 Your Digital Wellbeing

50. KEY TAKEAWAYS FOR STUDENTS

51. AI as your Copilot Think of generative AI tools as your helpful assistants. They follow your commands and perform tasks well, but it's up to you to use them wisely and responsibly. While AI tools can do a lot of things well, these tools can make mistakes because they are trained to always provide an answer, so it is important to stay alert. AI is Not Perfect Always Fact-Check Make fact-checking a habit. Do not blindly trust AI-generated information - always verify it with trusted sources to be sure.  Beware of Bias Generative AI models can sometimes show bias in their responses. Always ensure you review the outputs with a critical eye and be proactive by adjusting the prompts as necessary. Ensure that you give credit where it is due by always citing work that has been completed with the support of generative AI Always Cite Your Sources Do not share private information with untrusted websites or apps, and read privacy policies to understand how your data is used. Don’t forget you can use AI tools to summarize complex documents, but always remember to fact-check and verify!  Protect Your Information Communicating with an LLM that can appear to converse naturally with you can be very tricky. Establish healthy boundaries with technology by limiting screen time and spending time with your loved ones in real life.  Mind Your Wellbeing SECTION 4 Key Takeaways