Artificial intelligence is the simulation of human intelligence by computer systems.
This is Milo from RoboKind. Their goal is to make robotics and artificial intelligence accessible to all humans.
They create assistive technology, Milo is a facially-expressive social robot, used with an evidence-based curriculum for autistic learners. AI is all around us today, working on our phones, and other devices, the IoT items we use, maps, facial recognition, voice recognition, are the things we may be aware of, but there is much more.
AI became possible because of three main developments:
- Big Data developed the effective collection of large quantities of information and faster processing power.
- Algorithm development of stacking neural nets, allowed AI to learn better and faster.
- Graphic Processing Units (GPU) became cheaper, so the devices were more accessible.
These developments allowed for AI to progress and develop as quickly as it has.
AI is now the overarching descriptor for devices or machines that act in ways that are tech-smart.
Machine learning (ML) is a type of AI (a subset of AI).
Deep learning (DL) is a subset of ML.
Note that Machine Learning and Deep Learning are aspects of AI, however AI isn’t necessarily ML or DL.
As much as AI is useful and is making technology more efficient, we should remember that there are risks and potentially dangerous aspects to be considered, such as
- job automation (the loss of large numbers of jobs),
- potential malicious use of artificial intelligence: invasion of privacy (e.g. privacy-surveillance, facial recognition, data-hacking), digital security risk (e.g. criminals training machines to hack or socially engineer victims or superhuman performance), physical security (e.g. weaponising consumer drones, humanoid robots), or political security and misinformation(e.g. repression, or automated and targeted disinformation campaigns)
- autonomous weapons or potential AI arms race.
Be aware of all the potential risks and dangers, and make your learners aware of them as well. Stephen Hawking in 2017 stated "Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks". Elon Musk expressed the opinion that "AI is far more dangerous than nukes". Discuss the importance of digital ethics, and being responsible digital creators. If we are teaching our learners to code, we need to explain the potential power they hold, and discuss how it might be used. Like everything in life we have a choice! As teachers and role-models we have a responsibility to our learners and communities to have complex discussions and create awareness.
Exposing our learners to ML and AI
Let's explore some ideas:
Eliza Chatbot: Created in 1964, Eliza is one of the first chatbots ever created. The user types in a question or statement and Eliza will give you a reply. The replies are limited; it is fun and safe to chat. It could be useful for younger learners to build reading and writing skills.
Eliza can be found here https://tinyurl.com/AIEx-Eliza
MakeBlock AI extension: MakeBlock makes a range of robots. My favourite is their original robot mBot. MakeBlock now has an AI extension, which includes image recognition, speech recognition, text recognition, human body recognition, and natural language processing https://www.mblock.cc/doc/en/use-extensions/AI.html
Scratch AI extension: This is a space where projects can be created and saved if they use neural networks, Artificial Intelligence and/or Machine Learning https://scratch.mit.edu/studios/2924464
Coding with Chrome: Learn, improve, or teach coding skills in the Chrome browser. Users are able to code using Blockly, Coffeescript, HTML, Javascript and Python with output to Logo Turtle and/or connected robot-toys such as the Sphero, SPRK+, mBot and Lego Mindstorms. Explore AI machine learning kits. https://chrome.google.com/webstore/detail/coding-with-chrome/becloognjehhioodmnimnehjcibkloed
AI experiments with Google
Learners could play with projects others have made, or they could create their own projects.
Explore teachable machine https://experiments.withgoogle.com/teachable-machine
or 'AI and Drawing', 'AI and Music' and other experiments.
AWS educate - Amazon's cloud courses for learners. AWS (Amazon web services) https://aws.amazon.com/education/awseducate/14-and-older/ Self paced courses (14-18 year olds): Cloud Explorer Badge, Cloud Inventor Badge, Cloud Builder Badge and Advanced Options
IBM - Machine Learning for Kids, a variety of hands-on activities that introduce simple machine learning models to students through games and interactive projects https://www.ibm.org/activities/machine-learning-for-kids
Machine Learning for kids: https://machinelearningforkids.co.uk/
Teach a computer to play a game.
Uses: Scratch and App Inventor http://appinventor.mit.edu/
Introduction to Machine Learning (Google Applied Digital Skills) Make inferences and recommendations using data, train a computer, and consider ethical implications of machine learning.
https://applieddigitalskills.withgoogle.com/c/middle-and-high-school/en/introduction-to-machine-learning/overview.html
Machine Learning for Kids and Beginners (Udemy Course).
Learn Machine Learning to create powerful AI for Real-World applications! For Kids and Beginners, Parents and Teachers! https://www.udemy.com/course/machine-learning-for-kids-and-beginners/
Google Cloud - Hey, Student Complete this four-course series in order. The series is designed for students with little to no experience in cloud computing and is an ideal foundation for diverse career paths. Build your cloud skills and demonstrate your proficiency by earning badges along the way. https://go.qwiklabs.com/hey-student
And so the journey continues, expose your learners to the new technologies, while at the same time teaching them ethical responsibility to become truly responsible coders, programmers, and digital citizens.
Have fun!
Karen Walstra works with schools on educational change, professional development and strategic thinking. Contact us to engage.
A great overview Karen!! If people talk about Data Science, they tend to think of AI only. Description, exploration and modelling of data using Statistics is the other big branch. It is sometimes called Analytics, which then includes Statistics and Operational Research.