Jake Miller:
That's what's really cool about AI is it's helping teachers be more efficient. It's helping administrators be more efficient, but it's also helping students have their needs met.

Speaker 2:
That was Jake Miller, a personalized learning and ed tech specialist at Summit ESC and ESC of the Western Reserve. On this episode of the conversation, Rachel and Christina dive into the real world impact of AI in the classroom, exploring how both students and teachers can harness its potential for learning and innovation. Let's dive in.

Music:
(Music).

Christina Grady-Watts:
Hello and welcome back to the conversation where we aim to spark ideas and inspire change. I'm Christina Grady-Watts filling in for John Hambrick. Today, Rachel and I are excited to sit down with Jake Miller, author, keynote speaker, podcaster and personalized learning specialist to discuss AI and its role in schools. Jake, thank you so much for being here with us today.

Jake Miller:
Thank you for having me, Christina and Rachel. I'm excited. I'm excited to join you guys.

Christina Grady-Watts:
Yeah, we are too. Let's start by hearing a little bit about your background. What brought you to this point in your professional journey and what drew you to the intersection of AI and education?

Jake Miller:
So first off, I'm in Northeast Ohio. I am based out of Kent is where I live, and I've been in education now for 20, I don't know how many years. I don't like to count it because it makes me feel old. And started off as a middle school teacher, middle school math mostly, and then science and STEM. But what I found was I really was excited about educational technology and then I found that what I was actually really excited about was supporting teachers. So it was supporting teachers with educational technology most of the time, but really it's just the act of supporting teachers. So now in my role as a personalized learning and tech specialist, I get to do that a lot.
And when AI started to really, I mean it's been around for decades, but when it started to really come out in full force two-ish years ago, I just realized that's what they needed support with. So I was I think just as nervous and anxious about it as everybody else was when AI started to become really prominent. But I just went like, well, this is what's impacting education now, and the teachers need my help and so I'm going to try to help them. And that's just what I do. Yeah. I wish I could say because I love it and sometimes I love it, but it's just the helping of the teachers for sure.

Rachel Daniels:
Jake, I'm curious to hear, as you work with educators across many districts, what trends are you noticing in how AI is being used in schools?

Jake Miller:
It's cool because when it started out two-ish years ago when ChatGPT came on the scene, it was just like, go try out this cool tool and tell it to write a song about you or write a story about you. And people were just using it for silly stuff and now it's really pivoted to people are really using it to do their jobs or their learning or whatever it might be, more efficiently and more effectively, which is I think really exciting. If you talk to a large group of teachers now, most of them are using it in some way to impact the way they do things and so are the administrators and so are the students. So teachers are using it to help them make better resources, better instructional things, more differentiated materials, meet the needs of all of their learners, provide that universally designed curriculum for their learners because they've always wanted to do those things and they couldn't.
And now AI makes it possible that they could take their teacher expertise and do the thinking and let AI do the doing. That's what a lot of people like to say. You do the thinking, let AI do the doing. So they might be using tools like Diffit or Brisk, which are two of my favorite ones, to like, okay, here's an article we're going to use we're talking about. I've selected this using my teacher brain and understanding what's best for kids and my content area. This is what's best. Now I need this to be made at different reading levels or in different languages or for learners with a variety of different abilities, and they pop it into a tool like Diffit or Brisk, and it does that work for them. So it makes them more efficient. It does the doing for them, the things we didn't use to have time for.
And then you see on the next part is how's the learner using it. So the learner is then able to access different things based on what the teacher provided to them. But then there are student-facing AI tools now too. So students most can't use ChatGPT, for example, because 13 years old and up, and often doesn't meet the requirements of schools' privacy and data privacy kind of stuff. So a lot of schools, maybe they're not using ChatGPT, but they're able to use tools like MagicSchool, which has the MagicStudent platform or SchoolAI. The cool thing about both of those is that it kind of scaffolds kids into using AI because we need to start exposing them to it, unfortunately. We can't stick our heads in the sand about this. It's part of our world now. It's part of their future so we've got to help them learn how to use it.
We can't just, "Oh, they'll figure it out after they're out of school." No, that's not going to go well. So we've got to help them. So tools like SchoolAI and MagicStudent and MagicSchool are nice because the teacher curates what they're going to experience in there, like you could use these different features in here. And then they get a lot of data back afterwards. They could see the interaction the student had with the AI, so then they can also make sure students are using it appropriately. So it kind of provides that safety net there. And the goal there is to meet the needs of every learner. So that's what's really cool about AI is it's helping teachers be more efficient, it's helping administrators be more efficient, but it's also helping students have their needs met and see that as a normal thing, to use tools to support them in being successful.

Rachel Daniels:
Well, your previous comment about teacher interaction with AI reminds me of a reluctant group of teachers I was supporting, and I was explaining to them how they could take some of the heaviness off of their own plate by using some prescriptive AI tools. And a teacher, you could literally see a light bulb go on over her head and she bursts out to the whole room, "Hey guys, this just gets us to the students and to the data quicker. Why wouldn't we use this?" And I thought, that's great. That should be a T-shirt, right? It gets us to the students and the work quicker, yeah.

Jake Miller:
Well, yeah. It frees us up to do the work that the AI can't do, right?

Rachel Daniels:
Exactly.

Jake Miller:
And that's the work that we should be doing. And oftentimes it's the work that... I know in my classroom all of my years in teaching, it's the kind of things I wasn't able to do oftentimes because I didn't have the time, I didn't have the bandwidth to do it. So the one-on-one conferencing, the small group instruction, the working with kids on specific issues, the building of rapport, sometimes you don't have time for all of those really important steps or the looking at data, to your point. And the AI could take some things kind of off of our plate so that we can do that. So that's a really great astute point by that teacher.

Christina Grady-Watts:
So Jake, let's dig in a little bit to personalized learning. And can we just start with level setting maybe and telling us, just defining personalized learning?

Jake Miller:
So what I always tell teachers about personalized learning... Christina, you and I have known each other for a few years and you know I'm a story and anecdotes person. Stories, anecdotes and metaphors are how I operate. So the way I now talk about personalized learning is, about once a week we have a salad night at our house. My wife and I, we have three kids. And on salad night, we set out on the counter a bag of lettuce, a bag of spinach. I cut up some tomatoes and some cucumbers. We set out some olives, a couple different salad dressings, a couple kinds of cheeses, some shredded chicken and some pepperoni. And then we say to our kids, go ahead and make yourself a salad. And we have to provide some oversight to make sure they provide a good salad and make themselves a good salad of an appropriate size to get a good meal out of it. But otherwise they're empowered to do it themselves.
And so that's kind of a really simplified way of how I think about personalized learning. It's that the teacher does the curation of making sure the pedagogy and the content is sound, but we're empowering the learners to make their own salad, I like to say. So I've done some things. We know my daughter, for example, doesn't eat red meat, so I can't just put out just pepperoni for the protein options. I've got to put out the shredded chicken. So I've been intentional about what I do, but I don't tell my son who loves red meat, you can't put chicken on your salad. He wants some chicken too, so he might put some chicken and some pepperoni. So it's me differentiating, understanding the variability of my kids at dinnertime or my wife understand the variability of the kids at the dinnertime, but then empowering them to make that decision.
And so that's kind of how I think about personalized learning is empowering the kids to make some decisions on their path, their version of success. What's the best way for them to be successful and have a good salad, let's say. When you think about the fact that we put out spinach and lettuce because my daughter prefers spinach. We put out Ranch and Italian dressing. I can't eat things with eggs in them, so I have to have the Italian dressing. We put out chicken and pepperoni because my daughter doesn't eat red meat. We're understanding the barriers in the salad options. And so we're making sure that it's designed so those barriers aren't there. And I think AI can do that for us too. There's so many ways that if we think about the barriers that might present themselves in a learning environment or content or curricula, AI can help us overcome those barriers. Does that make sense?

Christina Grady-Watts:
It does. And I love that analogy. And you are always using the food analogy. I love the analogy. It's always about food with you.

Jake Miller:
It really is.

Christina Grady-Watts:
So I like that you just touched on that intersection of personalized learning and AI, which I kind of want to dive into in a little bit. I've just reading a lot from futurists and the trends that they're predicting in the future, and I keep trying to dig into, well, what impact will this have on the future of education? How is that going to change education? And I keep reading about it's coming back to personalized learning. So I'm wondering how do you see AI playing a role in personalized learning?

Jake Miller:
So the first and foremost reason, and the one that's already here, is just the universal design for learning, a kind of proponent of it where we think about what's necessary for some or essential for some they say in [inaudible 00:10:34] a lot. Essential for some, but beneficial for all. So for example, in a tool like MagicSchool, you've got the ability to have the text leveled to different levels. And so if kids can access texts at different levels or read aloud or talk to the built-in Chatbot to get support. Or if we could use Diffit to translate them or put them at different levels or use other tools to make them more relevant to kids. So that's kind of the front line stuff that's kind of already here, is that we can make different options for learners to learn from and empower them to choose what they need.
I think what's coming more in the future is seeing the curricula and the learning experiences be a little more catered to our students. I think one of the problems we see in contemporary schools is that school is becoming less and less relevant to kids. I think about when I was in school, I was asking my teachers a lot or wanted to ask my teachers a lot, why should I learn this? I think nowadays that's even amplified especially when kids have TikTok mindsets of how they process information, in one minute little bursts. And so things really need to be relevant to them and connected to the things they want to be maybe when they grow up or the ways they learn best. And so we're starting to see some technologies that have the ability to individualize curriculum and things like that for learners. A lot of schools in Ohio are working with a company called SchoolJoy and SchoolJoy actually crafts individualized curriculum for learners based on, these are the learning standards. Now for Christina, these are her interests and it's going to craft it to your interests.
And for Rachel, these are her interests. So it's going to craft it to your interests. And I think the thing that always makes me nervous is we need to make sure there's quality control with stuff like that. But still, I think that's something we're going to see more and more of where the curriculum really is directed to the learner based on the learner's interests and needs and things like that. And I think we're going to see some crazy things. And then I don't even know what else. I mean, you think about where we were two years ago to where we are now with the things AI could do, yeah, it's going to be crazy.

Rachel Daniels:
I feel like the natural shift in this conversation then is about teacher preparation and making sure that teachers are equipped with the knowledge that they need to use the right tools in the right time. And so I'm curious how you view the role of professional development meant in ensuring that educators feel confident to use AI tools, that educators feel empowered and they don't feel as though AI is in some ways saying that the art of teaching doesn't have a place. I mean, that's sort of where my head goes. So I'd love to hear your vision of professional development in the context of AI.

Jake Miller:
I think what I've learned over the years with professional development is as much as I like to be at the front of the room talking and telling these metaphors about food and restaurants and things like that, what really needs to happen most in professional development is teachers having time to try things out and teachers having time to explore them and teachers having time to set things up. So I think that's the most important piece of good professional development is just time. And the second most important part is examples. So much when we talk about personalized learning, for example, teachers are like, "Well, Jake, this sounds great, and I understand the whole idea of feeding your kids different salads, but what does that look like in my classroom?" And they want to see it. And so I think the same thing is true with AI. They're like, "What does this really look like?"
So I think one of the best ways to make that happen is teachers sharing with one another. Teachers saying, "This is how I used MagicSchool. This is how I used ChatGPT. This is how I used Diffit to make my job easier." So I think the two big pieces are seeing examples of it in use and having time to explore it and use it.
And then the third thing I'd put in there too, because this is so overwhelming and scary for a lot of teachers... I feel overwhelmed by AI sometimes and I'm an EdTech guy. This is what I'm into and I'm overwhelmed by it. So I think they just need that direct support. And I think for a lot of teachers that might look like instructional coaching. I think instructional coaching is a really, really strong opportunity for them to grow and become comfortable with something that they're scared about. And so if we have instructional coaches in place that also know about AI, I think there's a lot of potential there to support our teachers.

Rachel Daniels:
And I also wonder, we so often talk about differentiation for students, but even as a professional developer myself, I'm often delivering one size fits all professional learning to folks. And I am a strong proponent of really developing pedagogical content knowledge and giving math teachers math PD so they can become stronger math experts and so forth. So that's another way in which I would like to see the use of AI sort of taken, is helping folks in my capacity think about how can we leverage tools to make pedagogical content knowledge an important component for teacher development and moving outside of what we currently see as the PD landscape.

Jake Miller:
Yeah. And that's a great point because AI is beneficial for us as professional development providers in the same way that it's beneficial for teachers as content providers to their students. So we can go like, I'm providing a PD on this skill, and I have a group of science teachers in there and use the AI to support us in crafting different versions of what we're sharing based on their needs. And what I like to do is I find that the AI, I think, I'm sure everybody knows this by now, but AI likes to just pretend it's right about everything and just spit out information that may or may not be accurate.
I coach my kids' basketball team and I tried one time to say, "Okay, I've got to develop who's going to play in each quarter. I've got this many kids. Each kid needs to play this many quarters. These kids play these different positions" and I put it in. I'm like, "Go ahead and make me a rotation." And I go like, "Here's a rotation. It's perfect." And it would have one kid playing two positions in the same quarter. And I'm like, "ChatGPT, that's one person. You can't do that." And it's like, "Oh, sorry, let me try again." And it does it again and it's the same thing again. And I'm like, "No, you're not doing well."
So when I think of doing things, especially serious things like professional development or crafting lessons for a class or something like that, I like to give it as much information as possible. So I'm giving it articles and I'm explaining things and I'm putting all kinds of stuff in there so it's not inventing things, so that it's using things that I trust. So I'm giving it maybe content standards and maybe some pedagogical knowledge and all kinds of, as much content as I can to help it give me really good stuff.

Rachel Daniels:
Yeah, I agree. I mean, what you get from AI is only as good as the prompt that you're able to provide.

Jake Miller:
Yeah, for sure.

Rachel Daniels:
And that is so critical and the feedback. Folks have watched over my shoulder as I've been teaching people how to do particular things, and they'll notice that my Chatbot is my friend. I'll say, "Thank you so much for your contribution, but here's where you fell short." They're like, "Why are you talking to it like that?" I'm like, "I want it to be my friend." Oh, this is just such a rich conversation.
Jake, let's end here. Let's end with a call to action. So for educators who might feel overwhelmed or unsure about integrating AI into their teaching, what's the best way for them to get started? Are there specific tools or resources that you would recommend them exploring first?

Jake Miller:
I think the first and most important thing is you don't have to do it all at once. Just take a small step and then take another small step after that. Take another small step after that. I like to talk about how we use the term baby steps a lot, and when we say baby steps, what we mean is small steps. But I think the other part that we're missing there is when a baby learns to walk, they take small steps, but they also fall down a lot and then they get right back up. So when I think about baby steps, it's two things. It's take small steps, one step at a time, but it's also like be prepared for something to not work out right and be ready to get back up and keep going. So I think the baby steps is important to think about that.
So think about what's something where you could leverage AI? What's a painful part of your maybe lesson planning process or content creation or a need you have in school for students? So let's say for example, you have an article that's really important that kids read in your class next week, but you know it's above the Lexile level of a quarter of the kids in your class and you don't have time to make it fit that need, but it'd be really great if you could, so use AI to do that. And that's your first step. So you identified something that's a challenge for you right now, and then that could be your first way you use AI.
In terms of tools to use. I think even though ChatGPT is the most well-known one and one of the most capable ones, it could be a little bit overwhelming because you just go into it and you're staring at this black screen with a blinking cursor and you're like, what do I tell it to do? So I kind of like tools like Diffit or MagicSchool to start with just because like, oh yeah, here's a text leveler. Click on it and tell it the grade level I want this text and give it the link. So it's already ready for you. It's already built for you. And so those are a great kind of entry-level tool. And then you could keep using them forever too, but they're a great way to start off is with those simpler tools like that.

Rachel Daniels:
Yeah, I love that approach. Start small, but also that realization that this won't be without its pitfalls, and so be ready to jump back up again and keep the work moving forward. Well, Jake, we are so thankful for your time today, and it was incredible to just dip into your expertise for these few short moments. We expect that we'll be talking to you more and more as AI continues to grow and expand. And as I remind people, don't be like those folks back in the good old days when the internet was new, who said, "I'm not learning that thing. It's not here to stay." Right?

Jake Miller:
Yeah.

Rachel Daniels:
Embrace it and learn about it and move forward. Right? Thank you to our listening audience for tuning in today. We certainly appreciate your presence with us. For Christina, this is Rachel. Take some baby steps with AI, stay curious and keep the conversation going.