Digital subscriptions
Presentation
Homework:
View handout and webpage
Create a free account on ChatGPT
Try out some of the prompts
Share any useful prompts that you found or created
Zoom classroom: https://sdccd-edu.zoom.us/j/9191959460?pwd=OXh0RE9ZTVZTWElTMUQ0ZzAxQzExdz09
Passcode: emeritus
Objectives of the Class:
By the end of this class, we will:
Understand what AI is and how it's developed
Learn effective prompt engineering techniques relevant to their daily needs
Gain hands-on experience using ChatGPT
Develop confidence in applying AI tools in your everyday lives
AI is technology that enables computers to perform tasks that typically require human intelligence, such as understanding language, recognizing patterns, and making decisions.
• Smartphones and virtual assistants: When you ask Siri or Alexa a question, they understand and respond. AI can even distinguish between two voices (Our Alexa device recognizes us as Kim and Mary).
• Email filters: AI is used to filter out spam emails. Over time, it learns what is junk mail and what is important. It learns this with your own interactions as well as using algorithms to predict spam.
• Navigation apps: AI is used to provide the best driving directions, analyzing traffic patterns, road closures and even the time of day to suggest the best route.
• Online shopping and recommendations: When shopping online, websites may offer recommended products based on what you have ordered before or what others order when they purchase this item.
• Healthcare devices: Products like heart monitors and wearable fitness trackers use AI to analyze things like heart rate and activity. In some cases, they may be able to alert you to potential issues.
• Social media: Sites like Facebook use AI to show you things that you might be interested in, based on what you have interacted with before. They use your connections and your likes to predict who you might want to follow. And of course, they offer friend suggestions based on friends you have (pulling in data you have provided such as education history, previous locations, pets, interests and more).
Using AI is like teaching a computer to think and learn. Imagine AI as a smart robot that can do tasks, solve problems, and even talk like us, depending on what it has been taught.
AI uses vast amounts of data, recognizing patterns and making decisions. It includes the use of:
Machine learning (where the system learns from the data it is given)
Deep learning (where the system learns in an unsupervised way from unstructured data, learning to use predictions to provide answers)
Reinforcement learning (where the system makes decisions based on actions, and is evaluated along the way).
Five Best Use Cases for Machine Learning (ML)
(when data is structured and smaller in size)
1.Spam Detection in Email
ML algorithms like Naive Bayes or Decision Trees work well with labeled data (spam or not spam).
2.Credit Scoring & Loan Approval
Banks use structured data (income, debt, age, etc.) to decide whether to approve loans.
3.Customer Churn Prediction
Businesses predict if a customer is likely to leave using structured customer behavior data.
4.House Price Prediction
Based on features like location, square footage, number of bedrooms—clearly structured inputs.
5.Market Basket Analysis
Retailers analyze purchase history to recommend items (e.g., "Customers who bought this also bought...").
Five Best Use Cases for Deep Learning (DL)
(when data is unstructured and large in size)
1.Image Recognition & Classification
DL models like Convolutional Neural Networks (CNNs) can identify objects, faces, or animals in images.
2.Speech Recognition
DL (Recurrent Neural Networks or Transformers) can turn spoken words into text, like Siri or Alexa.
3.Natural Language Processing (NLP)
.Deep learning powers chatbots and translators, such as ChatGPT or Google Translate.
4.Self-Driving Cars
DL processes camera feeds, sensor data, and maps to make driving decisions in real time.
5.Medical Image Diagnosis
Deep learning can analyze X-rays or MRIs to detect conditions like tumors more accurately than traditional ML.
Data Collection: AI systems like ChatGPT are fed massive amounts of information—books, websites, conversations, and articles—similar to how you might learn by reading many books throughout your life.
Pattern Recognition: Just as you learned to recognize handwriting styles or your grandchildren's voices, AI identifies patterns in data. It notices that certain words often appear together or that specific questions typically have similar answers.
Neural Networks: These are inspired by our own brains. Think of it like a complex web of connections, similar to how your brain formed connections over years of experience. When you see an apple, you instantly recognize it because of these connections. AI builds similar pathways, but with mathematical calculations.
Learning From Feedback: Remember learning a skill where someone corrected your mistakes? AI improves the same way. When people indicate if an AI response was helpful or not, the system adjusts to do better next time—just like you learned from feedback throughout your life.
Language is the key to teaching AI. We teach AI by showing it lots of text, which helps it learn to make sense of words and sentences like a human does. This way, AI can understand and respond to us better. As we keep teaching AI with more language, it gets smarter and can talk or write in ways that are more helpful and interesting.
AI can be incredibly helpful, but it raises questions about how personal information is collected, stored, and used. Sometimes AI tools collect data about us without us knowing—through apps, smart devices, or online behavior.
Ethical use of AI means using it in ways that are fair, transparent, and respectful of people’s rights. It is also important to think about who is in control of the technology and who might be affected by it.
Companies and governments must make choices about how AI is designed and used—but we, as users, also have a role.
Prompts are the instructions or questions asked of an AI like ChatGPT to get a response (or output). The prompt can be a question, a statement, or a command. Simple prompts will create simple answers, but a thoughtful prompt may produce a much more meaningful example.
Creating good prompts is a skill which takes practice. But it helps if you can follow some steps:
1. Be specific and clear
o Instead of: "Help me with my medication"
o Try: "Create a simple daily schedule for taking 3 medications: lisinopril at 8am with food, metformin at 8am and 6pm with meals, and atorvastatin at 8pm."
2. Provide relevant context
o Instead of: "Tell me about gardening"
o Try: "I'm 75 years old with mild arthritis and live in a warm climate. Suggest low-maintenance plants and ergonomic gardening techniques that would work well for me."
3. State your purpose
o Instead of: "Tell me about smartphones"
o Try: "I recently got my first smartphone at age 70. Explain the basic functions I need to know in simple terms, focusing on making calls, texting, and using the camera."
4. Set the level of detail
o Instead of: "Explain Medicare"
o Try: "Explain Medicare Parts A, B, C, and D in simple terms. Use short paragraphs and avoid technical jargon."
5. Ask for formatting that works for you
o "Please format your response in large, readable text with bullet points."
o "Please break down this explanation into numbered steps."
Health: "Explain how to manage Type 2 diabetes through diet in simple terms. Include 5 easy meal ideas that are diabetic-friendly."
Technology: "Write step-by-step instructions for setting up video calling on an iPad. Assume I have no prior experience with iPads."
Leisure: "Suggest 10 books similar to 'The Thursday Murder Club' by Richard Osman, which features older protagonists solving mysteries."
Daily life: "Create a weekly meal plan for one person that minimizes cooking time and uses ingredients that stay fresh for several days."
An AI application is a tool or system that uses artificial intelligence to perform tasks usually requiring human intelligence. This can include recognizing speech, making decisions, translating languages, or identifying objects in images. Some examples of AI applications that you are currently using include things like virtual assistants, facial recognition, email filtering, customer service chatbots, health and fitness trackers and more.
In the image on the left, you can see many examples of AI applications.
Today, we will focus on ChatGPT, one of the more popular AI applications.
Go to the site (chatgpt.com)
Create a free accounts or use without an account
See below for more information of each level of use
·Basic conversations
·Help with creative writing and brainstorming
·Math and problem solving
·Summaries of articles, stories and more
·Recommendations for books, movies, music or others
·Learning support
·Access to GPT-4o mini and limited access to GPT-4
·Standard voice mode
·Limited access to file uploads, advanced data analysis and image generation
·Use custom GPTs
·Extended limits on messaging, file uploads and image generation
·Standard and advanced voice mode
·Access to deep research and multiple reasoning models
·Create and use tasks, projects and custom GPTs
·Limited access to Sora video
• "Write an email to my grandchild thanking them for their visit last weekend. Mention how much I enjoyed our walk in the park and the board game we played."
• "Help me plan a small birthday celebration for my friend who is turning 75. We have a budget of $100 and 5-6 people will attend."
• "Create a packing checklist for a 5-day trip to [location] in [season]. I need to pack medications and prefer casual clothing."
• "Explain how cloud computing works in terms a senior with basic computer knowledge would understand."
• "I want to learn about [topic of interest]. Create a beginner's guide with the most important concepts explained simply."
• "Compare and contrast streaming services like Netflix, Hulu, and Amazon Prime. Which might be best for someone who enjoys classic movies and documentaries?"
• "Write a short poem about the wisdom that comes with aging."
• "Suggest five new hobbies I might enjoy as a retiree who loves nature and history."
• "Create a short story about an older adult who discovers a hidden talent."
There are more AI tools than ChatGPT. Here is a comparison of some of the more popular AI assistants. Note there are many more AI assistants out there.
1. For everyday questions and creative help: Try ChatGPT
2. For simplifying complicated documents: Try Claude
3. For converting text to audio: Try Google Gemini
4. For researching current topics: Try Perplexity
• All these tools are available online through your web browser
• Free versions have some limitations but are still very useful
• Always verify important information from multiple sources
• Never share personal financial or medical details with AI tools
• Practice makes perfect – the more you use them, the better you'll get!
Learn about ChatGPT
Or, use the tool itself to find more. Go to ChatGPT. Ask it what it is, and how it works. You can ask if it saves your conversations. You can also ask if it shares information.