All posts by Ben Garside

Localising AI education: Adapting Experience AI for global impact

Post Syndicated from Ben Garside original https://www.raspberrypi.org/blog/localising-ai-education-adapting-experience-ai-resources/

It’s been almost a year since we launched our first set of Experience AI resources in the UK, and we’re now working with partner organisations to bring AI literacy to teachers and students all over the world.

Developed by the Raspberry Pi Foundation and Google DeepMind, Experience AI provides everything that teachers need to confidently deliver engaging lessons that will inspire and educate young people about AI and the role that it could play in their lives.

Over the past six months we have been working with partners in Canada, Kenya, Malaysia, and Romania to create bespoke localised versions of the Experience AI resources. Here is what we’ve learned in the process.

Creating culturally relevant resources

The Experience AI Lessons address a variety of real-world contexts to support the concepts being taught. Including real-world contexts in teaching is a pedagogical strategy we at the Raspberry Pi Foundation call “making concrete”. This strategy significantly enhances the learning experience for learners because it bridges the gap between theoretical knowledge and practical application. 

Three learners and an educator do a physical computing activity.

The initial aim of Experience AI was for the resources to be used in UK schools. While we put particular emphasis on using culturally relevant pedagogy to make the resources relatable to learners from backgrounds that are underrepresented in the tech industry, the contexts we included in them were for UK learners. As many of the resource writers and contributors were also based in the UK, we also unavoidably brought our own lived experiences and unintentional biases to our design thinking.

Therefore, when we began thinking about how to adapt the resources for schools in other countries, we knew we needed to make sure that we didn’t just convert what we had created into different languages. Instead we focused on localisation.

Educators doing an activity about networks using a piece of string.

Localisation goes beyond translating resources into a different language. For example in educational resources, the real-world contexts used to make concrete the concepts being taught need to be culturally relevant, accessible, and engaging for students in a specific place. In properly localised resources, these contexts have been adapted to provide educators with a more relatable and effective learning experience that resonates with the students’ everyday lives and cultural background.

Working with partners on localisation

Recognising our UK-focused design process, we made sure that we made no assumptions during localisation. We worked with partner organisations in the four countries — Digital Moment, Tech Kidz Africa, Penang Science Cluster, and Asociația Techsoup — drawing on their expertise regarding their educational context and the real-world examples that would resonate with young people in their countries.

Participants on a video call.
A video call with educators in Kenya.

We asked our partners to look through each of the Experience AI resources and point out the things that they thought needed to change. We then worked with them to find alternative contexts that would resonate with their students, whilst ensuring the resources’ intended learning objectives would still be met.

Spotlight on localisation for Kenya

Tech Kidz Africa, our partner in Kenya, challenged some of the assumptions we had made when writing the original resources.

An Experience AI lesson plan in English and Swahili.
An Experience AI resource in English and Swahili.

Relevant applications of AI technology

Tech Kidz Africa wanted the contexts in the lessons to not just be relatable to their students, but also to demonstrate real-world uses of AI applications that could make a difference in learners’ communities. They highlighted that as agriculture is the largest contributor to the Kenyan economy, there was an opportunity to use this as a key theme for making the Experience AI lessons more culturally relevant. 

This conversation with Tech Kidz Africa led us to identify a real-world use case where farmers in Kenya were using an AI application that identifies disease in crops and provides advice on which pesticides to use. This helped the farmers to increase their crop yields.

Training an AI model to classify healthy and unhealthy cassava plant photos.
Training an AI model to classify healthy and unhealthy cassava plant photos.

We included this example when we adapted an activity where students explore the use of AI for “computer vision”. A Google DeepMind research engineer, who is one of the General Chairs of the Deep Learning Indaba, recommended a data set of images of healthy and diseased cassava crops (1). We were therefore able to include an activity where students build their own machine learning models to solve this real-world problem for themselves.

Access to technology

While designing the original set of Experience AI resources, we made the assumption that the vast majority of students in UK classrooms have access to computers connected to the internet. This is not the case in Kenya; neither is it the case in many other countries across the world. Therefore, while we localised the Experience AI resources with our Kenyan partner, we made sure that the resources allow students to achieve the same learning outcomes whether or not they have access to internet-connected computers.

An AI classroom discussion activity.
An Experience AI activity related to farming.

Assuming teachers in Kenya are able to download files in advance of lessons, we added “unplugged” options to activities where needed, as well as videos that can be played offline instead of being streamed on an internet-connected device.

What we’ve learned

The work with our first four Experience AI partners has given us with lots of localisation learnings, which we will use as we continue to expand the programme with more partners across the globe:

  • Cultural specificity: We gained insight into which contexts are not appropriate for non-UK schools, and which contexts all our partners found relevant. 
  • Importance of local experts: We know we need to make sure we involve not just people who live in a country, but people who have a wealth of experience of working with learners and understand what is relevant to them. 
  • Adaptation vs standardisation: We have learned about the balance between adapting resources and maintaining the same progression of learning across the Experience AI resources. 

Throughout this process we have also reflected on the design principles for our resources and the choices we can make while we create more Experience AI materials in order to make them more amenable to localisation. 

Join us as an Experience AI partner

We are very grateful to our partners for collaborating with us to localise the Experience AI resources. Thank you to Digital Moment, Tech Kidz Africa, Penang Science Cluster, and Asociația Techsoup.

We now have the tools to create resources that support a truly global community to access Experience AI in a way that resonates with them. If you’re interested in joining us as a partner, you can register your interest here.

(1) The cassava data set was published open source by Ernest Mwebaze, Timnit Gebru, Andrea Frome, Solomon Nsumba, and Jeremy Tusubira. Read their research paper about it here.

The post Localising AI education: Adapting Experience AI for global impact appeared first on Raspberry Pi Foundation.

Experience AI: Teach about AI, chatbots, and biology

Post Syndicated from Ben Garside original https://www.raspberrypi.org/blog/experience-ai-new-updated-lessons/

New artificial intelligence (AI) tools have had a profound impact on many areas of our lives in the past twelve months, including on education. Teachers and schools have been exploring how AI tools can transform their work, and how they can teach their learners about this rapidly developing technology. As enabling all schools and teachers to help their learners understand computing and digital technologies is part of our mission, we’ve been working hard to support educators with high-quality, free teaching resources about AI through Experience AI, our learning programme in partnership with Google DeepMind.

""

In this article, we take you through the updates we’ve made to the Experience AI Lessons based on teachers’ feedback, reveal two new lessons on large language models (LLMs) and biology, and give you the chance to shape the future of the Experience AI programme. 

Updated lessons based on your feedback

In April we launched the first Experience AI Lessons as a unit of six lessons for secondary school students (ages 11 to 14, Key Stage 3) that gives you everything you need to teach AI, including lesson plans, slide decks, worksheets, and videos. Since the launch, we’ve worked closely with teachers and learners to make improvements to the lesson materials.

Access the Experience AI Lessons

The first big update you’ll see now is an additional project for students to do across Lesson 5 and Lesson 6. Before, students could choose between two projects to create their own machine learning model, either to classify data from the world’s oceans or to identify fake news. The new project we’ve added gives students the chance to use images to train a machine learning model to identify whether or not an item is biodegradable and therefore suitable to be put in a food waste bin.

Two teenagers sit at laptops and do coding activities.

Our second big update is a new set of teacher-focused videos that summarise each lesson and highlight possible talking points. We hope these videos will help you feel confident and ready to deliver the Experience AI Lessons to your learners.

A new lesson on large language models

As well as updating the six existing lessons, we’ve just released a new seventh lesson consisting of a set of activities to help students learn about the capabilities, opportunities, and downsides of LLMs, the models that AI chatbots are based on.

With the LLM lesson’s activities you can help your learners to:

  • Explore the purpose and functionality of LLMs and examine the critical aspect of trustworthiness of these models’ outputs
  • Examine the reasons why the output of LLMs may not always be reliable and understand that LLMs are machines that make predictions
  • Compare LLMs to other technologies to assess their suitability for different purposes
  • Evaluate the appropriateness of using LLMs in a variety of authentic scenarios
A slide from an Experience AI Lesson about large language models.
An example activity in our new LLM unit.

All Experience AI Lessons are designed to be cross-curricular, and for England-based teachers, the LLM lesson is particularly useful for teaching PSHE (Personal, Social, Health and Economic education).

The LLM lesson is designed as a set of five 10-minute activities, so you have the flexibility to teach the material as a single lesson or over a number of sessions. While we recommend that you teach the activities in the order they come, you can easily adapt them for your learners’ interests and needs. Feel free to take longer than our recommended time and have fun with them.

A new lesson on biology: AI for the Serengeti

We have also been working on an exciting new lesson to introduce AI to secondary school students (ages 11 to 14, Key Stage 3) in the biology classroom. This stand-alone lesson focuses on how AI can help conservationists with monitoring an ecosystem in the Serengeti.

Elephants in the Serengeti.

We worked alongside members of the Biology Education Research Group (BERG) at the UK’s Royal Society of Biology to make sure the lesson is relevant and accessible for Key Stage 3 teachers and their learners.

Register your interest if you would like to be one of the first teachers to try out this thought-provoking lesson.  

Webinars to support your teaching

If you want to use the Experience AI materials but would like more support, our new webinar series will help you. You will get your questions answered by the people who created the lessons. Our first webinar covered the six-lesson unit and you can watch the recording now:

September’s webinar: How to use Machine Learning for Kids

Join us to learn how to use Machine Learning for Kids (ML4K), a child-friendly tool for training AI models that is used for project work throughout the Experience AI Lessons. The September webinar will be with Dale Lane, who has spent his career developing AI technology and is the creator of ML4K.

Help shape the future of AI education

We need your feedback like a machine learning model needs data. Here are two ways you can share your thoughts:

  1. Fill in our form to tell us how you’ve used the Experience AI materials.
  2. Become part of our teacher feedback panel. We meet every half term, and our first session will be held mid-October. Email us to register your interest and we’ll be in touch.

To find out more about how you can use Experience AI to teach AI and machine learning to your learners this school year, visit the Experience AI website.

The post Experience AI: Teach about AI, chatbots, and biology appeared first on Raspberry Pi Foundation.

How anthropomorphism hinders AI education

Post Syndicated from Ben Garside original https://www.raspberrypi.org/blog/ai-education-anthropomorphism/

In the 1950s, Alan Turing explored the central question of artificial intelligence (AI). He thought that the original question, “Can machines think?”, would not provide useful answers because the terms “machine” and “think” are hard to define. Instead, he proposed changing the question to something more provable: “Can a computer imitate intelligent behaviour well enough to convince someone they are talking to a human?” This is commonly referred to as the Turing test.

It’s been hard to miss the newest generation of AI chatbots that companies have released over the last year. News articles and stories about them seem to be everywhere at the moment. So you may have heard of machine learning (ML) chatbots such as ChatGPT and LaMDA. These chatbots are advanced enough to have caused renewed discussions about the Turing Test and whether the chatbots are sentient.

Chatbots are not sentient

Without any knowledge of how people create such chatbots, it’s easy to imagine how someone might develop an incorrect mental model around these chatbots being living entities. With some awareness of Sci-Fi stories, you might even start to imagine what they could look like or associate a gender with them.

A person in front of a cloudy sky, seen through a refractive glass grid. Parts of the image are overlaid with a diagram of a neural network.
Image: Alan Warburton / © BBC / Better Images of AI / Quantified Human / CC BY 4.0

The reality is that these new chatbots are applications based on a large language model (LLM) — a type of machine learning model that has been trained with huge quantities of text, written by people and taken from places such as books and the internet, e.g. social media posts. An LLM predicts the probable order of combinations of words, a bit like the autocomplete function on a smartphone. Based on these probabilities, it can produce text outputs. LLM chatbots run on servers with huge amounts of computing power that people have built in data centres around the world.

Our AI education resources for young people

AI applications are often described as “black boxes” or “closed boxes”: they may be relatively easy to use, but it’s not as easy to understand how they work. We believe that it’s fundamentally important to help everyone, especially young people, to understand the potential of AI technologies and to open these closed boxes to understand how they actually work.

As always, we want to demystify digital technology for young people, to empower them to be thoughtful creators of technology and to make informed choices about how they engage with technology — rather than just being passive consumers.

That’s the goal we have in mind as we’re working on lesson resources to help teachers and other educators introduce KS3 students (ages 11 to 14) to AI and ML. We will release these Experience AI lessons very soon.

Why we avoid describing AI as human-like

Our researchers at the Raspberry Pi Computing Education Research Centre have started investigating the topic of AI and ML, including thinking deeply about how AI and ML applications are described to educators and learners.

To support learners to form accurate mental models of AI and ML, we believe it is important to avoid using words that can lead to learners developing misconceptions around machines being human-like in their abilities. That’s why ‘anthropomorphism’ is a term that comes up regularly in our conversations about the Experience AI lessons we are developing.

To anthropomorphise: “to show or treat an animal, god, or object as if it is human in appearance, character, or behaviour”

https://dictionary.cambridge.org/dictionary/english/anthropomorphize

Anthropomorphising AI in teaching materials might lead to learners believing that there is sentience or intention within AI applications. That misconception would distract learners from the fact that it is people who design AI applications and decide how they are used. It also risks reducing learners’ desire to take an active role in understanding AI applications, and in the design of future applications.

Examples of how anthropomorphism is misleading

Avoiding anthropomorphism helps young people to open the closed box of AI applications. Take the example of a smart speaker. It’s easy to describe a smart speaker’s functionality in anthropomorphic terms such as “it listens” or “it understands”. However, we think it’s more accurate and empowering to explain smart speakers as systems developed by people to process sound and carry out specific tasks. Rather than telling young people that a smart speaker “listens” and “understands”, it’s more accurate to say that the speaker receives input, processes the data, and produces an output. This language helps to distinguish how the device actually works from the illusion of a persona the speaker’s voice might conjure for learners.

Eight photos of the same tree taken at different times of the year, displayed in a grid. The final photo is highly pixelated. Groups of white blocks run across the grid from left to right, gradually becoming aligned.
Image: David Man & Tristan Ferne / Better Images of AI / Trees / CC BY 4.0

Another example is the use of AI in computer vision. ML models can, for example, be trained to identify when there is a dog or a cat in an image. An accurate ML model, on the surface, displays human-like behaviour. However, the model operates very differently to how a human might identify animals in images. Where humans would point to features such as whiskers and ear shapes, ML models process pixels in images to make predictions based on probabilities.

Better ways to describe AI

The Experience AI lesson resources we are developing introduce students to AI applications and teach them about the ML models that are used to power them. We have put a lot of work into thinking about the language we use in the lessons and the impact it might have on the emerging mental models of the young people (and their teachers) who will be engaging with our resources.

It’s not easy to avoid anthropomorphism while talking about AI, especially considering the industry standard language in the area: artificial intelligence, machine learning, computer vision, to name but a few examples. At the Foundation, we are still training ourselves not to anthropomorphise AI, and we take a little bit of pleasure in picking each other up on the odd slip-up.

Here are some suggestions to help you describe AI better:

Avoid using Instead use
Avoid using phrases such as “AI learns” or “AI/ML does” Use phrases such as “AI applications are designed to…” or “AI developers build applications that…”
Avoid words that describe the behaviour of people (e.g. see, look, recognise, create, make) Use system type words (e.g. detect, input, pattern match, generate, produce)
Avoid using AI/ML as a countable noun, e.g. “new artificial intelligences emerged in 2022” Refer to ‘AI/ML’ as a scientific discipline, similarly to how you use the term “biology”

The purpose of our AI education resources

If we are correct in our approach, then whether or not the young people who engage in Experience AI grow up to become AI developers, we will have helped them to become discerning users of AI technologies and to be more likely to see such products for what they are: data-driven applications and not sentient machines.

If you want to use the Experience AI lessons to teach your learners, please sign up to be the first to hear when we launch these resources.

The post How anthropomorphism hinders AI education appeared first on Raspberry Pi Foundation.