Strategy Beats the Tool in AI for Schools ft. Karle Delo | My EdTech Life 365

Welcome And Sponsor Thanks

Dr. Alfonso Mendoza 0:12

Hello everybody and welcome to another great episode of My Ed Tech Life. Thank you so much for joining us on this wonderful day. And as always, wherever it is that you're joining us from, thank you, thank you, thank you so much from the bottom of my heart for all of your supports, the likes, the shares, the follows. Thank you so much for engaging with our content. We really appreciate it. As you know, we do what we do for you to bring you some amazing conversations. And we would also love to thank our sponsors. Thank you so much to Comeback Coffee, as always, for keeping us caffeinated and energized to continue to do our podcast. Thank you so much to Book Creator, Eduaid, and Peelback Education for your support as well. And today I am really excited because I have a wonderful guest who's been on the show before when I did a panel show, and this was you know a couple of years back, and boy, time sure flies. And now I'm really excited to bring her back on the show that she's gonna tell us about a new role that she moved into, and she has some great insight because she works with so many districts, and so she gets to see what's good, what's getting better, and maybe some of the scary stuff. But it's just to get a great perspective of what's

Meet An AI Strategist

Dr. Alfonso Mendoza 1:30

happening for many of us that are within our districts. Sometimes we're siloed, we only see what's happening here in our local area. But just for you to hear what's happening elsewhere, this is something that is amazing. So I would love to welcome to the show my great friend and just a wonderful person overall, uh Carly Dilo. How are you doing today?

Karle Delo 1:51

Doing great, how are you?

Dr. Alfonso Mendoza 1:52

I am doing wonderful, Carly. Thank you so much for being here and again just sharing your perspective on things. So, Carly, just so we can go ahead and before actually we get started in and get into the meat of things for our audience members who are tuning in right now and are listening to you for the very first time today and getting to know who you are. Can you give us a little brief introduction and what your context is within the education space?

Karle Delo 2:19

Yes, thank you. So, right now I'm an AI strategist at Michigan Virtual, which is a statewide nonprofit, and it's a really cool job. It's really interesting because I do get to travel across the state and talk to different districts and educators and students about what they're doing with AI guidance, with AI implementation. So it's an amazing role. And last time I came on the show, I was in the role of curriculum director. So I've got that past history as well with over 14 years in public education. And so it's really interesting to be able to zoom out and see that perspective because I'll tell you that each district that I walk into and each building that I walk into has got a vibe, as the kids would say, or as we would call the culture, but it really is something distinct that you can feel and notice in those different areas. And so it's so interesting to see what the common threads are and then also what's unique about these different districts.

Dr. Alfonso Mendoza 3:11

I love it. And you know, and that's something that is so important, and I think here uh for as long as we've been doing the show, and obviously since um, what is it, November of 2022, you know, and and you've seen my show, you've engaged with our content, and just really growing into being that cautious advocate and yourself included too, just being, you know, with the experience that you have, you get to see so much. So I love that today people are gonna hear just a different voice, different perspectives. So I want to ask you, Carly, you know, transitioning into this role with uh, you know, now as an AI strategist, but before we get into that, you know, per se, I want to ask you, you know, how was that transition

Lessons From Curriculum Leadership

Dr. Alfonso Mendoza 3:52

like? You know, you working with a district, doing curriculum, and now moving into this AI strategist role. What are some of the lessons that you're taking that you took from your previous role, bringing it into this role, and maybe just kind of tweaking things and just learning along the way?

Karle Delo 4:10

Yeah, one thing that we did in my previous district is we really thought about how we can use AI to advance our existing initiatives and work towards what we're already doing. So it was a small rural district. We didn't have a ton of extra staff or extra time. So we figured out with our elementary staff, we were rolling out a new reading curriculum, K5. So we actually used Notebook LM to store some of those curriculum resources right off the bat. So the training wasn't about AI, it was about their new reading curriculum. But here's this amazing resource that we could utilize. And then we did similar things at the secondary level as well. We figured out what are we already trying to do? How can we use AI to help us get there? And we really started by focusing on making sure teachers understood the technology first before even mentioning or discussing student use. So we were already in spring of 2023 doing that kind of organically. So that was that was really interesting. And I'll also share that we we started a task force pretty early on to create guidance, but a lot of that was inspired from some of the things I've heard on on this show. So when I first started interacting with AI, me being someone who's highly interested in technology and always has been, I thought it was really cool. And then I started doing a lot more listening at that district leadership level. I realized this isn't about my opinion or what I think about AI. This is about what's best for the community and what's best for kids. So I've got to listen to other perspectives. I also listened to your show and and you had so many different types of voices on it to be able to see what are the risks and limitations as well as the benefits. And so that really shaped the the approach that I like to take and the more balanced, cautious advocate type of approach, but also really helping schools take a proactive approach because if we don't do anything, then we have all of this AI use that's happening in secret in pockets where people aren't talking about it. And that's where I think there's a huge risk. So those were some of the lessons taking over into this role. But then also it's just been it's a been a wild transition and really interesting because I do get to collaborate and hear the best of all these ideas from from all these different places. And I noticed that the more perspective that we gain and the more we hear from other people and are able to collaborate and connect people, that's when we get really great ideas and results coming together.

Dr. Alfonso Mendoza 6:37

Absolutely. And I think that's something that is wonderful that again, somebody in your role and what you're doing and just making those connections and seeing various school districts, the way they do things, and obviously because a lot has to really fall in line also with their own mission and vision of, you know, implementing AI or still kind of w waiting and seeing. But like you said, you know, many times not doing something, they're still doing something and not doing something, which is obviously missing out on maybe opportunities that might be there. And like you said, it's also about informing community. So I know that right now you're in a very unique position where you kind of really have a bird's eye view of kind of what's happening in all of these pockets of schools that you currently work with. So when

What Districts Ask For First

Dr. Alfonso Mendoza 7:25

a school district reaches out to you, usually what are some of the main questions that they are reaching out uh to you for or about?

Karle Delo 7:35

At this point, we are known for helping districts create guidance and a comprehensive implementation and integration plan. So that's typically what the ask is. Um, is they they want to know how can we get this work started at our district. And so it many times it does start with guidance. Some people already have guidance in place, and then they're asking us to come in and say, okay, we've got this document now. So those are some of the big components. And then people still want training as well, as far as like basic professional development and learning. So we can offer that in components, but we also have a network of AI ambassadors, for example, across the state where we could tap into those individuals as well to help with that. Um, but what we'll also do is help with targeted training. So we will go in and maybe help the admin team or the leadership team understand how it applies to their role specifically or the high school language arts department. So that's a lot more of the work we're doing as well. But what they want to, what people want to figure out is um what can we do about this as far as taking a common approach? Are there any things that we should are there things that we should ban? Are there things we should block? What about AI detectors? Are those okay to use? How can we get kids to stop cheating? I mean, those are some of the big questions that people have right off the bat. Um, and we want to help them navigate through all of that.

Dr. Alfonso Mendoza 8:59

Excellent. Now, for our audience members that are listening in, and even though we're already in 2026, you know, since 2022 and the release of Chad GPT, a lot has changed in our field, in the field of education, and our field is instructional technology, technology and everything as a whole. But there are still school districts out there that are still holding back. And in my experience, there are still some school districts, you know, nationwide that are still being a little uh well very conservative as far as AI and really the use of technology. So I want to ask you, and maybe for our audience members that are listening, because you know, we may have some of those decision makers that are listening from those states and so on, could you walk us through maybe some of the what you might do for them if they do partner with an entity like yours? And I'm not saying every entity is going to do things the same way, but should they be maybe in your area? How can a school partner up with you and what do you kind of walk them through maybe in those initial 30 days to kind of maybe introduce them, put them at peace and ease about maybe that guidance, those policies? How what's a good starting point for them?

A Smart Starting Point For AI

Karle Delo 10:18

The first thing I like to do is help demystify AI and teach the basics of what it is and how it works. And we usually do that through an interactive um exploration and then discussion as well. So we're in there, we're asking questions, we're leading with curiosity, but the goal is to help people figure out that this is not an all-knowing machine. It's not a magic wand, it's not this perfect um this perfect crystal ball that's gonna tell you what's gonna happen, like demystifying all of those components. Um, but at the same time, while we do that, we both see the benefits of the tool, but also the risks and limitations come out when we start to dig into that. And we have an honest conversation about how even though these risks and limitations exist, um, there's still there's still some power here as far as this as a learning tool, thinking about let's start by thinking about the adults. And so that's where we begin. And um one one example, one thing we like to share is you know, AI is not an all-knowing machine, but there's no all-knowing source of truth. Like Google isn't an all-knowing source of truth. There's no one person that is. So it doesn't mean that it's not a valuable tool. It just means we can't approach it that way. And what I find is that a lot of people are approaching it in that way. So that's the first component. And once teachers and educators have a better understanding of how the tool works, what it's good at, what it's not good at, then they are more prepared for that student conversation. So I always suggest that districts start there with building teacher and adult understanding. Um, and that also includes like when not to use it. So it's not just saying use it for everything, but have that filter of um, this is a good use case, this isn't, you know, this is not such a good use case. Um, and then from there, we really want to tie it to what are the values of the district and thinking about a common guiding principle, for example, would be um we want to use AI in ways that enhance and support teaching and learning, not replace it. So we think about what are the most key, like three to five really big values for this district. And then from there we'll go into specific do's and don'ts, and that's where the guidance component comes in and get feedback on it. So that's what the task force group will do. And it's just getting everybody at least on the same page as far as a starting point. Another thing I like to share is not everybody has to be an AI super user. And I think that's a mistake some people make is they're like, we got to do all this training and everybody's got to be able to do all these things. Not necessarily. You want a certain group of people in your district to know how to use it very, very well. And it helps if that person's also an instructional leader, because that's when the magic happens when they start to see of, oh, I could make a gem that would help the entire language arts department with this initiative we're already we're already started on. And that's when we get those components, but not everybody's got to be there. So at least raise the floor for everybody and then give some of those other people who are more of the early adopters and innovators, give them space to at least collaborate, learn, and grow. Um, and I wouldn't feel I don't think that that always involves more screen time or more technology for students or even using AI with students. I've seen a lot of powerful ways that teachers have used AI as a thought partner or throughout the planning process, but then students aren't even necessarily interacting with the AI. So that's something that that districts decide with what works with their current policies, their current technology plan, um, but it doesn't always mean more screen time. In fact, it's really interesting more than if there's a lot of those uh AI chatbots that are designed for education, for example. And I think there's a misconception that some people think we should just have students on those all the time, or if they're being used, they're used in that way. Really, like five to seven minutes is usually a good amount of time for a even with adults for any sort of like chatbot interaction. Um, so it can be overused, and we do not want that. Um, that was a big thing we saw it actually in some districts after COVID when people came back and teachers were used to having so much of their content online. So they just kind of like kept it there, even though students were back face to face. And in in our district at least, we made an we started being more intentional about pulling back some of that technology and focusing on human connection. And so the same is true here. We can't push just the technology or we're gonna end up with a similar problem. Uh, we have to talk about the strategies behind it.

Dr. Alfonso Mendoza 14:57

Oh, that's great. I mean, you you shared so much, but what I do appreciate though about this is that I think what you're saying too is uh there are already districts that at least have an idea of what they want to do, but they just require that additional help and support that you and your organization offer to at least just bring that vision to fruition for them, and because they they already at least know the path, that alignment, and they already have that, you know, in mind as to what they want this to look like where it's gonna benefit, like you mentioned, not only the teachers, but it's also gonna benefit the students, and it's about informing the community as well, you know, making sure that they are involved in knowing what is gonna be used, what isn't gonna be used, how it's going to be used, and I think that's something that is great. One thing that you did mention too is just that that importance of demystifying uh AI, and like you said, it does not know everything, although it sounds very confident in the answers that it gives. But as a matter of fact, I had a a guest, uh actually the most recent episode that I released, uh, and I I posted the clip today where she was saying, you know, I wanted to use a quote from one of my colleagues that I respect a lot. You know, she popped in information and then she got an output, and then she said, Well, I'm gonna make sure and double check this before I say anything and speak. Uh, and then of course it was a hallucination. The the citation was not correct, the quote never came directly from that specific colleague. And so she talked about the importance of your reputation, how it takes years to build, but one mistake like that can just, you know, the hammer falls on you. AI is not gonna be to blame, it's gonna be the person that's gonna be to blame. So I think it's just so many things there that obviously cause that fear. However, like you mentioned, taking small steps, having those conversations, what it can and can't do is something that is great to kind of just get those conversations going. And the chat bot, also very interesting that you brought that up because I'm pretty sure you saw it on the news and you know, everybody had been talking about it about Sal Khan saying, you know, Conmigo was a non-event, you know, and Conmigo is actually dying out. And if you've noticed too, even in Chat GPT, study mode is not visible, you know, in uh depending on the account that you have, and then you have to do like a forward slash in order to at least see it. So it's very weird how there was that big push of here's this chat bot, here's this, to provide that support, and now it's like okay, let's kind of reel it back and just bring it back, and just to kind of I guess either rethink the process, reformulate, you know, and just see what else will come out or what other innovation. So a lot of interesting things happening right now. But I want to ask you, you know, with your experience, if you wouldn't mind sharing, and obviously you don't have to we you don't disclose any names or school districts or anything, but I know that you have seen many districts that have had success implementing AI, but then you've also seen a lot of other districts that just really have fallen flat when it comes to this. So I want to ask you, Carly, in your experience, what is the difference between a district that is actually moving the needle and moving the needle forward, and one that ends up with an awesome policy, but yet nobody even looks at that paper and it's kind of still like the wild, wild west and may fall flat?

Why AI Rollouts Stall Out

Karle Delo 18:38

Well, I'll say um one mistake across the board that I see districts making or schools making in general is when it comes to thinking about the professional development behind AI, and a lot of them um do like a intro, like AI introductory type of half day or full day training, and then they call it good. Um so a lot of districts don't even get to that guidance piece. Um, another mistake I'll see people make is just they'll do the stoplight model and say this is our AI policy, which there's that's that's not gonna work because of a lot of different reasons. Um but going into that that first professional development component, um it just a one and done PD is not gonna solve the problem. Um, and what it takes is really people doing what I talked about earlier with finding how does it fit into existing initiatives, and then you bring it into that. So if, for example, the professional learning is about um making sure that your assessments are standards aligned, maybe there's um a component where teachers are interacting with a gem or AI using a specific prompt to run their assessments through to see how standards aligned they are or to get feedback. So that's one example of how it can be done really successfully. And another common mistake that I see going back to that PD component is a lot of people who provide PD on AI and technology are are somewhat tech enthusiasts. That's why they're in the job and they forget that their audience is a range of people from tech minimalists all the way to up to enthusiasts. And so that's where we see like, here's 20 tools here, and that's way too many. Like I usually stop at three. So 20 is usually too many for people, and it's just like a demo heavy, um like type of magic show. Like, here's here's all the amazing things it can do without going into what the risks and limitations are and how much time it actually should take to vet and how to vet those things. What are the red flags to look for? And I'll tell you that the balanced approach to PD, it is less exciting. It is, it's less exciting than the the wow, fast-paced demos, but it is more realistic and it paints a clear picture of what it can do and what it can't do. Um, so that is the approach I recommend and it's got to be sustained over time. Um, so bring thinking about how could if if you're talking about it and you've got, let's bring in the guidance document. Um, the guidance document should be part of that professional learning. If you have guiding principles, it should be anchored in that. Like those should be the different sections of your PD, or you should be referencing back to it constantly. But in addition to that, what if um at a staff meeting, the principal is also talking about um, here's one way you could use this tool to work towards what we're already doing. This aligns with XYZ, or maybe even saying, you know, we shouldn't put, remember, if you use AI for this, don't include student data, make sure you anonymize data, things like that. So those are things that districts do that are successful, and also where I see a lot of those misses, especially when it comes to PD. So people have got to keep that in mind. Um, and also a lot of districts that are very successful, they'll they'll take the all of this as a starting point and then run with it. So, one thing I like to do at the end of uh any sort of meeting or Professional learning is ask people, what did you get out of this session and what do you want moving forward? I work, I've worked with one district who to who took those wants and they they created quick shops that were like 30-minute after school um little workshops where people could attend optionally and they were addressing exactly what teachers said that they wanted. So that was that was really incredible, and that's a great way to sustain this over time. And then next year, that district is going into more choice-based sessions, but they also had um their coaching team get get training, like anyone who provided PD, we we did a workshop together where they started to think about okay, I'm coaching people on formative assessment. How could AI support that initiative? And so it can't just always come from one person, it can't always come from an outside expert. Um, it should, you know, it's great to have a variety of different voices that are in that and also try to find ways that you can create an internal group of um early adopters, innovators who are also sharing what they're finding success in across the district. Um, so those are those are some different, I mean, I've got so many stories so that I I don't want to throw it all at you at once. Um, but those are some high-level, high-level things that I'm seeing.

Dr. Alfonso Mendoza 23:22

Well, that's great. Those are some great shares. And you know, one thing I did want to share too is like you mentioned, you know, the the tech enthusiasts, and just like ourselves too. I mean, we go to a lot of conferences and so on, and I I just love the way that you refer to it because it's like it's a magic show, you know, kind of like here's what these 20 apps can do. But in reality, you're you like you said, there is no depth to that professional development. Now, I'm not saying there isn't a time and a place to do kind of those app smashes, those quick, like, hey, 20 tools you can use tomorrow and things of that sort, you know, little bites, uh, bite-sized PDs that you can just, you know, do. But I think that you're right. And in a lot of cases, is it just feels like like, okay, we already checked the box, we got our one and done PD, so everybody's ready, you guys are good, all right. Let's go, and we're all in full on. But that's not the way it works, and like you said, it it there's a care that goes in it, like there's a cycle, you're checking up on people, you're seeing where they're going, you're seeing the progression. I might add that I I did partake in an event uh that happened last week, and it was really neat that I went into one of those sessions, and the presenter said, Well, and they were sharing how they use excuse me, AI in uh their district as an instructional coach. However, she provided AI and non-AI exemplars because she said, Well, you know, in our district, we still have some teachers that do not or choose not to use AI in the classroom. But I thought that that was such a great thing for an instructional coach to, like you said, going back to those tech enthusiasts. Sometimes we don't know our audience because we get so excited. But I love that they still offered that non-AI approach and still created where, okay, here for the AI user, here's a gem that you can use that's going to provide you this information. For the non-AI user, here is a template, an outline, and here are some suggestions of how you might be able to go through this. And I was like, you know, we can't forget that there are still some teachers that are still a little hesitant about that, and it's it's it's okay. It's natural to still feel that way. It's natural to feel like, hey, you know what, I don't want to depend on this, or I don't want this to, you know, just really take over a lot of things, whatever the case is. But I found that to be very novel and really getting to know your audience, who it is that you have in your district, and still making the or still giving them opportunities to continue to grow and learn. And eventually, I think, like my great friend Renee Dawson said, you know, in every district you have those speedboats, those tugboats, and those anchors. So work with the speedboats. That's going to be, like you said, that task force of people that are going to be very familiar with the tools that can easily turn around and help, whether it be their content level uh colleagues or maybe at the district level also as well. Then you have those tugboats that are kind of there in the middle, and then they'll kind of start pulling those anchors as once they start seeing, like, oh, okay, I see how this is you know making things a little bit easier for you in in certain ways. So, you know, it all depends on on your makeup there in your classroom, but and in your district, but let's not dismiss that too as well. So that's very important. Now, I want to ask you a a very I don't know if it's a silly question or anything, but you know, if you could put a billboard up, you know, on your main highway there in Michigan, where every morning teachers drive by, superintendents drive by, and instructional leaders drive by and see uh what would be the three biggest mistakes you would want them to read so that they may avoid when it comes to AI in their districts.

Karle Delo 27:16

All

Three Mistakes Leaders Must Avoid

Karle Delo 27:17

right, the top one is gonna be very practical, but there needs to be more conversation about it and more awareness, and it's stop putting student information into chat GPT or any personal AI account because I still get surprised faces when I share that information. So I think that that would be that'd be the top um choice there. The second one is that um, you know, there's there's an AI policy that's being built with or without you. So there's all there's this culture and there's AI use happening with or without intention. And I think people need to address that. I heard from I talked to a group of students just last week, and one of the students said a lot of adults just don't want to want to pretend like we it doesn't exist and like we're not using it, and that is so far from reality. So, and it also is with adults. Adults are using it with or without guidance as well in pockets, and a lot of times it's like this secretive thing, and we don't want that. So that's the second one. And then just the third one is strategy beats the tool every single time. It's not about the tech tool, it's about the strategy. I can use um AI to make a lot of busy work and worksheets and multiple choice quizzes and questions, you know, I can use it to automate those things, or I can use it in really powerful ways to bring all of those what-if ideas that teachers have had to life as far as like project-based learning or um getting students to really be creative throughout their their process and showing what they can what they can make, what they can learn. So there's so many different possibilities, and it's the strategy that beats the tool every time.

Dr. Alfonso Mendoza 28:53

I love that. That's a great billboard for all those leaders to look at. And I I do want to say that it is very important, and even now, you know, four years, going into four years now since uh Chad GPT and all these large language models, but you're absolutely right. It it as far as education leaders putting in that PII, that personal identifiable information, into these large language models to, you know, IEP reports, 504s. Um, you know, there there was a an incident where there was uh you know a staff member that was using these large language models, and I was like, why are we getting errors? You know, why are we getting these errors? Because it it's a it's a platform that feeds into another platform so we can get some information. And all of a sudden, when we opened up that file, we saw kind of like a lot of uh the garbage or the artifacts, and what was happening is that that member was copying and pasting straight from Chat GPT, but it had emojis, and they were copying and pasting all of that content into that platform, and the other platform could not make out what that was, and it was you know just we weren't getting accurate information and data until we found out later on we put in a work ticket, and they're like, Oh, we're seeing a lot of emojis, we're seeing this, this is why it's breaking, and so that kind of gave it away because if you notice, there was a time where everything had emojis on it, and so you know it's very dangerous, and it's very important that we still see um that the importance of the data privacy aspect of it, like you mentioned, you know. So I want to ask you just kind of a nice segue into this question as adults.

AI Slop And Trust Breakdown

Dr. Alfonso Mendoza 30:40

I know we've talked about, and yeah, I know you've talked a little bit, and we hear it on the news too, about AI slop showing up, not only just on the student side, but with the grown-ups. You know, we're talking about educators, we're talking about administrators. What does AI slop look like coming in from the teachers and administrators, and why is this a bigger problem than most people think?

Karle Delo 31:07

Great question. So uh going on the emoji reference, if you're if your stuff still has the emojis directly out of Chat GPT, it is definitely AI Slop. And I have seen resumes turned in that have the emojis and all, and that is a huge problem. It's a huge red flag, so just just don't do that. Um, but thinking about the bigger problem that it creates, AI Slop is any content that has little to no value. There's usually a lot of fluff in it, and a lot of people um think about that and and associate it with different AI videos or images that are out there, but it also appears a lot in different writing. I mean, if you get if somebody sends me a five-paragraph email that doesn't say much, doesn't have a lot of value for me as the reader, but has a lot of fluff, that's AI slop communication. Um, and so part of it is that as humans, when we receive communication from somebody, for it to have meaning, usually it means someone had to spend some time thinking about crafting that message. And so there's a there's a level of importance to that, and it's something we should keep in mind. The golden rule that I share with people about AI for communication, if you're gonna use AI to help at all with communication, it needs to be not only beneficial for you, the person sending the information, but also the receiver, the person receiving. So an example of how I might use it is let's say I'm writing an email, it gets really long, and I realize um that I'm all over the place. I might use AI to revise it for clarity, make it more clear, make it easy to scan. And one thing, especially early on, that I tried to do was I would ask people and tell them I used AI to help with this email, was it useful for you? And so it has to have value on both ends of it. Another time that I think is a good way to use it is to maybe summarize meeting notes or action steps or things like that because I can get that out more quickly to people. And I would feel comfortable telling them, hey, I used AI as a support here. Now, if it's a message that I'm crafting be like in reply to somebody who um where I haven't even fully thought through how I want to approach my response, it doesn't make sense to use AI there. I should sit down because writing is the thought process. So I need to sit down and think through it and actually write that out. So really, we just don't want to turn to it as the first thing all the time. And what we really don't want is to create this cycle of AI slop where AI is just communicating with itself. So, for example, let's say teachers start generating all their work with AI and then students are using AI to do the work, they turn it in, and then teachers use AI to grade it, and it all just becomes this AI cycle. That is not what we want. That's not the goal. It also can really break down trust over time because if if you're just always sending something and not putting a lot of thought into it when it comes to communication or or even things that you're handing out to students, like students, they are using these tools, they know and can recognize that. So if you hand out something that you didn't revise, edit, or put a lot of thought into, they might pick that up as AI slop. And and a really interesting argument that I hear is that like busy work is also technically like some sort of like maybe human slop. Like that's something that has existed before, and that's true, but AI can automate it and make it so much quicker, which is why it can be a problem. Um, one story, this is from a friend who is an AI advocate. Um, so it's something she speaks about, and her her daughter um got a got some feedback on an essay, and she recognized right away, like her daughter was reading through it, and she's like, This makes no sense. Her mom looked at it and noticed right away, this is Chat GPT, this is AI generated feedback, and it's wrong. So she emailed the teacher this very nice message of like, here's what I noticed, and um, I think that you used AI for this, and that's okay. AI is a lot of great purpose, you know, went into this whole reasoning and rationale. The response she got from the teacher was like literally a copy and paste from Chat GPT. So she couldn't even communicate or contact the teacher, and it's like that that's not what we want. That's not the goal. And so I think we need to be really cautious and careful. And the other thing is if anybody can use AI to generate writing, which oftentimes is very mediocre on its own, um, if if there's not the right prompting or revision, it just becomes noise at some point, and so that's it, you know, we want to be able to teach students that to be able to stand out and to stand out, it's actually gonna require your authentic voice. Be in a world of AI slop, authentic voice becomes so much more powerful because it breaks, it actually breaks the pattern, it doesn't blend into everything else. So I will take an unpolished human response um over AI generated fluff any day.

What Students Really Think

Dr. Alfonso Mendoza 36:15

Very interesting. I love that. Now you hit on something there talking about the students and talking about the feedback, and I know in your role, you get to work with a lot of districts and work right alongside students too as well. I mean, you're there at the district. So I want to ask you in your experience, Carly, what is it right now that the students are saying about AI and AI in education? What what's the what's the big thing that resonates?

Karle Delo 36:44

So consistently, some of the most interesting and um you know, qu some of the most interesting conversations that have pushed my thinking the furthest have actually been with students. And a lot of that's because their perspective is unique and outside of where where we're thinking as adults. And so their voice and their opinion is really important here. And what I hear is that they are thinking about this really deeply on a deep level and having more nuanced conversations than most adults give them credit for. So a lot of people think that just all students are using it to to just cheat and do their work for them. Um, but there's a lot of other interesting stuff that's happening as well. On the topic of AI grading, this was really interesting. This came up last week. I asked some students about this, and they they were high school students, and they're like, Yeah, we it feels like a slap in the face when teachers use AI to give feedback, and they especially if they don't like share that that's what they're doing or explain why. And their response was, I could do that myself, and I probably can use AI better than my teacher. So they're thinking I could get better feed, like I could have done that myself. They want human opinion and human feedback. On the flip side of that, I've also worked with a teacher who had who set up a chat bot where students used this chat bot to get feedback, and then they conferenced with the teacher on which feedback to take or not. That that's a really powerful positive use because students were part of the decision-making process and there was transparency involved. So it's just interesting that they all unanimously felt like that breaks down trust if teachers are using it in that way. Um, but they are concerned about AI diminishing their critical thinking skills just like adults are. I hear that in conversations, and we see that in national survey data as well. And they are using it in many cases in some really interesting ways, and they don't always know if it's okay or not. They're looking for guidance from adults. And right now they're learning from places like YouTube, TikTok, um, and each other, maybe they're parents. Uh, but when I ask that question of where do you learn about this, they very rarely say that it's from adults or from teachers. So this is a really big opportunity where we are able to potentially shape how these young people interact with and think about these tools. And we're we're missing out if we're not at least talking about it. So there's there's teaching about AI and teaching with AI. Um, and it we at least need to be teaching about or having that conversation. So that's that's a big part of it. And they they just have a lot to offer. A moment that stands out to me, I was um moderating a student panel discussion where the audience was full of adults and leaders from schools, and the kids were all talking about how they used AI to quiz themselves. And I'm like, so you all do this, and they're like, Yes, we've all done this. And I turned around and I asked the adults, have any of you used AI to quiz yourselves? And nobody raised their hand. So it's just really interesting how these students are using it as this learning tool in ways that adults, you know, we're more like we're the ones more likely to use it to just write something for us, you know. Um, so it's so it's just interesting and they have a lot to offer to the conversation.

Dr. Alfonso Mendoza 40:09

Yeah, actually, you know, I I think you hit on something very important there that I really like and kind of takes me back to the classroom where, you know, oftentimes as teachers we feel like we have to be the know-it-alls and we need to know the subject, we're the subject matter experts in the classroom and so on. But, you know, I think that I obviously I didn't grow up grow up to be a great teacher. I learned how to be a great teacher, but I learned most from my students. I mean, somebody that did I like myself, I didn't go through the traditional education route, you know, through the university, you know, I'm coming in from business and marketing, and then uh getting, you know, an opportunity to get into a math classroom just because I had enough math credits with my finance uh uh you know aspect and coming in. And really, I mean, you know the basics of the teaching, but I mean really it was the learning process from the students, learning process from my colleagues, and it was just being that learner and understanding that it's okay to not know everything, it's okay that you were not the expert, and it's okay to be vulnerable. But in these cases, I think right now, you know, being able to have that open dialogue within within your classroom with students, providing discourse, providing feedback, going back and forth, I think that's such a great learning experience for the teacher as well, because now you they get to see, like, oh, I never thought I can use it for that, or I never thought I can use it for that, where students, like you said, are being innovative and taking charge of their learning and how they're using it. Yeah, obviously, everybody assumes that they're just gonna go home, copy, paste, and that's it. Now, I I'm not saying that they don't do that, there are some that do, but however, there are others that really take it to that next level. So it's important, like you said, to have that open communication, talk to one another, build that community, and learn from one another. And it's amazing how much I learned from my students when I was in the classroom, so much so that I think they were like some of the they were the catalyst for me being able to move into the district level as a digital learning coordinator because I was able to take risks with them. I was able we were able to try things with them, we were able to just, you know, have fun and enjoy the learning and understanding also like failing forward is a thing. Like, I mean, if it didn't work the first time, it'll work that next time. But I think uh, you know, one of the things there that really stuck with me is that having those conversations, learning what their perspectives are, because sometimes we as adults we just think like, hey, nope, what I think is right, and I'm just closed off and I'm just siloed here with my colleagues, so we're absolutely right. But there's a generation here that's right in front of us that we can still continue to learn from and learn from one another, so I really like that. So it's important having that discourse in the classroom wonderful.