Artificial Intelligence and searching
Searching The Web - Lesson 6
Objectives
- To understand and explain artificial intelligence (AI) in simple terms
- To create clear and precise prompts for a generative AI chatbot
- To understand possible negative uses of generative AI
Lesson Resources
- Lesson Slides
- Whiteboards and pens or notebooks for the students
- Access to a generative AI chatbot account (e.g. ChatGPT, Copilot, Google Gemini) for the teacher only.
Lesson 1 - Refined Searching
Lesson 2 - Other ways of searching
Lesson 3 - Evaluating websites 1
Lesson 4 - Evaluating websites 2
Lesson 5 - Searching Maps
Lesson 6 - Artificial Intelligence and searching
Before the lesson
Later in the lesson you demonstrate a generative AI tool to the class. We have used the example of ChatGPT, but you could also use other similar systems such as Microsoft Copilot or Google Gemini. Ensure you have an account ready for use before the lesson and that the system is available to access on your school network.
Introduction
The way we search the online world is rapidly changing. You may have heard about artificial intelligence or AI and the ways that it can be used to search for, or create things online. But what is AI and how can it help people?
Put the class into small discussion groups and ask them to try and answer these two starter questions:
- What do you think AI is?
- When and where do people use it?
Collect answers and examples from groups. They may mention popular generative AI websites such as ChatGPT or Copilot, or they may understand that smart assistants such as Apple’s Siri and Amazon’s Alexa make use of AI Machine Learning.
Show the class this video to introduce some basics of the subject area:
What is Artificial Intelligence for Kids | What is AI | AI for Kids | AI explained for Kids |AI Kids
This definition from IBM, sums it up well:
“AI is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy”
AI has been around for a long time in different forms. People have dreamed of robot assistants in science fiction since the 1920’s and actual applications of 'intelligent' machines have been around since the 1950's. Here's some other key events and dates in the progression of AI:
- 1952 - Arthur Samuel developed the first computer ‘checkers’ (draughts) playing program and the first computer program to learn on its own.
- 1988 - Rollo Carpenter develops the chat-bot Jabberwacky to "simulate natural human chat in an interesting, entertaining and humorous manner." It is an early attempt at creating artificial intelligence through human interaction.
- 1995 - Cliff Kushler, co-founder of Tegic invents T9, "Text on 9 keys." The predictive text technology displays words with a single key press.
- 1997- Deep Blue becomes the first computer chess-playing program to beat a reigning world chess champion.
- 1998 - Dave Hampton and Caleb Chung create Furby, the first domestic or pet robot.
- 2000 - Honda's ASIMO robot, an artificially intelligent humanoid robot, is able to walk as fast as a human, delivering trays to customers in a restaurant setting.
- 2009 - Google starts developing, in secret, a driverless car. In 2014, it became the first to pass, in Nevada, a U.S. state self-driving test.
- 2014- Amazon develop their virtual assistant Artificial Intelligence system, Alexa. Alexa can play music, provide information, deliver news and sports scores, tell the weather condition, and control your smart home.
- 2020 - Chat GPT-3 is released. This version is based on natural language (NLP), deep learning, and Open AI, enabling it to create sentence patterns, not just human language text. It can also produce text summaries and perhaps even program code automatically.
Credit - Forbes - A short history of AI and Verloop.io - The Timeline of Artificial Intelligence – From the 1940s to the 2020s
So people have been imagining the possibilities of AI for a long time, but the last few years have seen big strides in the technology and it being more available and accessible to everyone through the many smart devices in our lives.
AI is a widely used term that actually covers a wide range of uses. But let’s put the focus back on searching and creating content, which is known as generative AI.
Generate means to create or produce something, let's take a look at this video to explain.
Generative AI explained in 2 minutes
Let’s try an example of generative AI together together!
At this point, its important to make it clear that most generative AI tools require a user account to be created, and the vast majority require the account holder to be 13+, so this is not an activity you can get your students to do themselves, with their own accounts. But you can create your own account and demonstrate its use in class.
Use with care and responsibility:
- Be sure to check that the systems you wish to access are available in your school and carefully consider your prompts you will use in the lesson (examples are given below).
- Rehearse the activity before doing it live with the class
- Freeze your screen or hide your window from the screen that students can see when prompting the chatbot.
- Follow your school’s online safety policy at all times.
Using a generative AI Chatbot
Let’s open ChatGPT, which is a large language model chatbot. GPT stands for Generative Pre-Trained Transformer.
The requests we give it are called prompts. The clearer and more precise our prompts the better the results will be.
Its called a chatbot as its designed to feel like you are having a conversation with someone and they are providing the things you have asked for. You can go back and forth and ask extra questions to add more detail to the prompt and change, improve or modify the results.
Let’s give it this prompt:
> Please create a nursery rhyme in the style of 'Incy Wincy Spider' but about a snail
Ask the class - What do you think of the results?
These were the results we received:
Here's a nursery rhyme about a snail, inspired by the style of "Incy Wincy Spider":
**Slippy Slowy Snail**
Slippy slowy snail,
Crawled up the garden path,
Down came the rain,
And gave the plants a bath.
Out came the sun,
And dried up all the dew,
So slippy slowy snail,
Crawled back to munch some stew!
This rhyme keeps the rhythm and playful nature of "Incy Wincy Spider," while featuring a snail in a garden setting.
Let’s add some extra detail to the prompt. What else could we want to include in our new nursery rhyme? Add these new prompts to gradually modify the results. Again, the results we received are shown in italics each time. Your results may vary.
> The snail should live in a tree
Here’s a version where the snail lives in a tree:
**Slippy Slowy Snail**
Slippy slowy snail,
Climbed up the big oak tree,
Down came the rain,
As soggy as can be.
Out came the sun,
And dried the branches high,
So slippy slowy snail,
Went sliding to the sky!
Now the snail's home is in the tree, with a fun upward journey!
> Please make the snail meet a spider during the rhyme
Here’s the rhyme where the snail meets a spider:
**Slippy Slowy Snail**
Slippy slowy snail,
Climbed up the big oak tree,
There he met a spider,
Spinning fast and free.
Down came the rain,
And washed the web away,
So slippy slowy snail,
And spider went to play!
Now the snail and spider have a little adventure together in the tree!
> The snail should be called Steve
Here’s the rhyme with Steve the snail:
**Steve the Slowy Snail**
Steve the slowy snail,
Climbed up the big oak tree,
There he met a spider,
Spinning fast and free.
Down came the rain,
And washed the web away,
So Steve the slowy snail,
And spider went to play!
Now Steve the snail has a name, adding a friendly touch to his adventure with the spider!
>Please draw a picture of Steve
Here is a drawing of Steve the snail, climbing a tree and meeting a friendly spider! Let me know if you'd like any changes.
> Please make Steve the snail wear a red baseball cap with 'Steve' written on the front.
Here is Steve the snail, now sporting his red baseball hat with his name on it! Let me know if you'd like any other details added.
Ask the class - Are we happy with the results?
You could keep tweaking the results for as long as you like, but let’s go back to that first prompt:
> Please create a nursery rhyme in the style of 'Incy Wincy Spider' but about a snail
What were the key words that made the results successful?
- Nursery rhyme
- In the style of...
- Incy Wincy Spider
- Snail
These all gave ChatGPT reference points for what we wanted.
Main task
Use paper or whiteboards and ask the children to work in groups and to come up with a brand new starter prompt for ChatGPT to create another (different) adapted nursery rhyme for the class.
Ask them to feedback their starter prompts.
Then ask them to pass their starter prompt to the group to their left. Ask all groups to try and add some more detail to the starter prompt.
Repeat another pass to the group to their left and adding another layer of detail.
Take the finished prompts and try a couple out, making sure all prompts are appropriate.
Generative AI and Large Language Models
So ChatGPT is an example of generative AI using a Large Language Model, let’s finish by finding out a bit more about how they work.
How Chatbots and Large Language Models Work
Ask the class - So have we searched or created using ChatGPT today?
Probably more created than searched, in terms of looking for something specific that is already out there. But the way ChatGPT and other Large Language Models work is by doing exactly that: looking at the data it has been trained on from what is already out there on the web and making predictions based on what it finds.
So whether its searching or creating, its certainly an amazing way of looking for what you need!
To finish
Ask the class to consider these final questions as a group, before feeding back their thoughts:
- Can you think of any negatives of this technology?
- Could it be misused in any way?
In terms of misuse, the main things to draw attention to are the potential for people creating AI generated content and claiming it as their own work or real when it is not. Whether this is writing, images, songs and sound files and even videos. The technology now has the ability to create really convincing 'fakes' that could mislead people into believing something is true when it isn't really. This could cause great upset and anger if someone is made to look like they are doing or saying something negative.
If we are going to make the most of the technology, it's vital that we understand the ethics of using it respectfully and for good.
BBC Bitesize regularly posts monthly real or AI quizzes for images and videos. See if your class can spot the AI fakes from the genuine content.
Their Other Side of the Story section is a great resources which is "Helping you navigate fake news and misinformation and be more critical and curious about what you see and share online."
The other, hidden negative of generative AI is the environmental impact it has. It uses huge amounts of electricity and computing power to run a generative AI sessions, which can impact our planet. A typical Chat GPT query uses roughly 10 x more electricity than a normal search engine search, so we should really be mindful of when and how we use these services. This short video touches on some of the main environmental impact issues: