Teaching Kids the Imitation Game: The Importance of Educating Students about Artificial Intelligence
Teaching Kids the Imitation Game: The Importance of Educating Students about Artificial Intelligence
Albert Lee knows a lot about education.
As an AI training engineer for Clinc, he spends his days helping grow the company’s artificial intelligence software through machine learning and training data. However, Albert doesn’t only help machines learn. He also devotes a lot of his time to teaching computer science and artificial intelligence to middle-schoolers.
Albert is currently teaching a virtual class for sixth through eighth graders that he developed in partnership with the Detroit Area Pre-College Engineering Program (DAPCEP), a nonprofit organization devoted to STEM education with more than 40 years of experience partnering with universities, training programs, and K-12 school systems. Albert says that the class is “a great opportunity for young students to experience machine learning and build cool applications.”
We recently sat down with Albert to talk about his educational outreach work and why it’s important to introduce kids to the concept of AI at a young age. Here’s what he told us:
Why do you think it’s important to get STEM students thinking about concepts like artificial intelligence in middle school?
Albert: The short answer is “Why not?” Artificial intelligence is a fun topic. But also, when you’re working with younger kids, it’s much easier to develop curiosity around the topic. When you start thinking about AI with older age groups, they can start focusing too much on the applications or how it applies to their futures or careers.
With young kids, we get a lot of wild questions, which is great. I really encourage my kids to explore and try to break things when they’re working on our projects. I don’t want them to be afraid to try things.
How do you teach a concept as big and complex as artificial intelligence to a sixth-grader?
Albert: What I really enjoy about teaching AI is that everything within artificial intelligence is actually just an imitation of actual intelligence, right? So, you get to have these really bright discussions about what human intelligence actually is and how we, as humans, understand the world around us.
One of the exercises I do with the kids is an animal identification project. Together, we work to build an image recognition model that can tell the difference between a jaguar and a leopard. I mean, they’re both cats with spots. So, we gather training data, we work to train a machine learning model. Even if you personally don’t know the difference between the cats, you can train a machine learning model to recognize that difference. It’s a really good activity to get kids to see the value in AI.
Another activity we do is an extremely simplified version of the conversational AI we develop at Clinc. I have our students create their own imaginary friend and, through gathering data and machine learning, we can actually train that virtual AI friend to answer basic questions over the phone.
What does it actually mean to “train” AI?
Albert: As a human, one of the things we do is we learn from our experience, right? For example, if you go to a bank teller, you know exactly what type of interactions to expect because you’ve been to a bank many times before, so you’ve learned from those past experiences. With machine learning, our AI models learn from experience, but that experience has been coded into what we call training data. So, instead of having to actually programmatically code all of those instructions, you train your AI model on what you want it to do by providing it with training data.
What I enjoy most about artificial intelligence goes back to that theme of AI as an imitation of humans. One of the cool things about Clinc’s conversational AI technology is how its built in the same way that you actually experience conversations. It’s aligned with how humans really interact in person and that’s a fun concept to convey to students who are trying to wrap their heads around the technology.
If your students could only walk away from your class with one major insight into AI, what would you want it to be?
Albert: I mean, first and foremost, I want them to understand that AI is fun and you can use it to build really cool stuff. AI is super interesting because it’s an imitation of what humans do and you can use it as a tool to create whatever you want to create. That’s a powerful thing for kids to understand and that’s why I love teaching classes like this.
But I’m not doing it on my own. Volunteers are so essential to these kinds of classes, particularly virtual classes, to help guide the students in breakout groups or projects. We have a lot of volunteers from Clinc working with my students, but we also have virtual volunteers from other companies and even other countries. I currently have one volunteer from the UK and another from South Africa. They’re very generous with their time.
There is a real community out there that is devoted to teaching kids about AI and I’m happy to be a part of it.
You can learn more about the programs offered by the Detroit Area Pre-College Engineering Program (DAPCEP) here.
You can learn more about how Clinc uses AI to create automated, human-inspired solutions for business and industry here.