audio1589817718

00:00:00 Speaker: Welcome back to the At Scale podcast. My name is Ben Sterling. I'm here with Tennessee Achieves. I'm so excited to be recording another podcast episode now in winter twenty twenty five. It's been a while since we recorded one gram. Good to be back. Yeah, I've been excited to be here today. And I'll tell you what I really like about this episode today is that this is our first listener suggested guest. So that means people are listening. Um, and if you're listening today and you've got a guest that you think, wow, they'd be really good on the podcast, uh, this is real life proof that we take those suggestions to heart. And we're here today to talk about a topic that I know is top of mind for everyone. I don't think, um, anything's made me happier than hearing that we had a recommendation off of a, um, an organic find of our podcast. So exciting for us. It's been all friends of the program up until this point, so it's great to bring in an external perspective. Um, today we have, uh, Jeff Riley from, um, day of AI, which is an organization. I'll allow him to do all of the intro of this, um, but an organization that's very committed to, uh, talking about the introduction of AI in the classroom and in education. Jeff has a wide variety of educational background, um, I think most recently serving as the commissioner of elementary and secondary ed in the, in the state of Massachusetts, which is a really cool job. Um, this one that you do now seems really cool. Um, so, Jeff, I'll let you to introduce yourself, talk a little bit about your, your background and kind of how you got to this unique point of your career. Sure. Well, first, thanks for having me, I appreciate it. Good to talk to you guys today. Yeah. So I had done thirty two years in public education. I started as a teacher. I think I did every job in between adjustment counselor, assistant principal, principal, deputy superintendent, superintendent and I ended my career as the commissioner of the state. Uh, which you say sounds like a cool job until Covid hits. Uh, and, uh, you know, after thirty two years, I was getting ready to retire. Retire. And, uh, our friends at MIT called me and they said, hey, we're launching a nonprofit out of MIT, uh, that focuses on AI and education, and we want you to help lead it. And I said, well, can I say what I want? And I think they got a little concerned with that question. And they're like, what do you mean? And I said, well, fifteen years ago we didn't regulate social media and cell phone use. And the research that's coming out now appears to show that, you know, we might have some mental health challenges with our kids because of that AI, despite the fact that I think it's going to be the next revolution in education, uh, also has some problems. So I want to be able to present the material when we talk to schools and districts in a fair and unbiased way where we show people the possibilities, but also the pitfalls that come with AI. Yeah, I, I that is a great perspective to have on this. I, I to um, kind of see this as a. major milestone in the education space. I we had computers when I was in elementary school, but it wasn't they weren't very complex and it wasn't too far before that where there were no computers in the school building. And I've heard from people about how against that, a lot of school districts were even just introducing a computer in the school building and then internet the school building and then now social media. And, um, it's another one of those I think, is, is that right? Is it kind of fit into that same sort of narrative? I mean, I think it's along those lines. I guess as a longtime educator, I'm a little cynical because technology has always been the silver bullet that's going to save. We just put a laptop in every kid's hands, and that's never been the case. Right? I actually think with AI, it may be the case because it's going to be able to help teachers make their lives easier, but also allow teachers to differentiate and personalize instruction in a way they never could before. So it's like fundamentally different than a lot of the technologies that have come in the past and that it can impact teacher pedagogy and practice. That's a great point. Yeah, it's it's definitely sometimes seems smarter than me when I use it, that's for sure. Um, yeah. So we work as an organization primarily with college students. So students that are transitioning from high school senior year into college. A lot of the students that, um, work with our program are at community colleges. Um, they're representing some sort of, um, vulnerable background. Eighty nine percent of our students represent a vulnerable background of some kind. Uh, and I think access to this, this type of technology can be a challenge for those students. And really understanding that we actually, uh, just conducted a survey of our students in about twelve thousand students responded to it, and seventy two percent of them said they don't use AI, which was really interesting to me. Um, because I think that number is going to have to go up, it's going to inevitably go up. Um, and so for as an organization who I guess we're interacting with a student population that's not really utilizing this tool yet. What advice do you have for us in terms of like trying to introduce that to that population of students? Yeah, I mean, I think what I speak about this publicly, I remind people that, you know, you are already using AI every time you try to send a text that it tries to predict your next word. That's artificial intelligence. Every time you're on Netflix and it tries to curate for you movies it thinks you will like based on your previous viewing history. That's artificial intelligence. We were in Arizona maybe six months ago, training all the state education commissioners in the country on AI. And we took a Waymo, a self-driving car. Freaky. But that's artificial intelligence. So you know. But so I think like, it's here, uh, we have seen some skepticism among students. Uh, and for a few reasons, I think I think, um, this is the first generation that saw what social media did, particularly to some of their peers and their mental health. So they're a little skeptical about that. Um, I also think there's a nervousness. This, particularly his kids transition from high school to college about what is fair and foul. When you use AI, am I going to be blamed for cheating if I use AI? And so people are a little bit including, uh, young people are a little bit skittish. And so I think the adults have to kind of iron out what the ground rules are for this. Yeah. Jeff, I like, uh. Well, first you said that you took the job only if you're allowed to say anything that you like. We like guests like that. So, uh, definitely say what's on the top of your mind here today. But, yes, I like how you started from an educator perspective. I really was kind of approaching this today and thinking like a student perspective. But I guess this question could go either way from as an educator, as a student, the three of us on the pod. Uh, that it's here. I haven't really summer. We went on a trip a little bit to help put together an itinerary, but maybe that's not used. Maybe, uh, our students that are listening for the beginning of school years with faculties who have a range of opinions. And, you know, again, some people are very nervous about using it. And my advice is pretty simple. Like, we're not going to be able to hold back the tide of AI. We're not going to be able to bury our heads in the sand. We are already using AI. You are. You may not even know it, but you're already doing it. And my recommendation is we start surfing this wave. And when I think about it, I say, you actually don't need to surf it, you just need to dip your toe in the water. And then I lay out for teachers how AI can make their lives easier and unburden them from a lot of the administrative handcuffs that we put on them, so that, God forbid, they could go back to focusing on building great, exciting lessons, and kids are more engaged in their learning because we've just, you know, we've handcuffed our teachers with a lot of administrative minutia and burdens, and AI can take a lot of that off their plates. Totally. We're doing that here. Tennessee achieves I mean, I would say a lot of our administrative stuff that we've spent a lot of time working on. We figured out ways to use some of the AI tools to really speed a lot of that up. Yeah. I mean, one of the things I show when I, when I speak with teachers and I train them is, you know, there's never been a teacher who hasn't had to write a note home to a parent when maybe the student wasn't behaving as well as they should. And parents and teachers will wring their hands trying to figure out, how do I word this correctly? You can explain to ChatGPT what you want. Johnny was misbehaving today. I need to be strong about it, but I don't want to offend the parent and suggest a path forward and within 10s. It can give you a nice outline of a letter. That's pretty good. And you know that maybe you need to tweak still. Maybe. But we'll have saved you hours of hand-wringing about how do I word this? And what do I say? And, you know, how do you do this? And it's those kind of moments where teachers are like, wow, that can actually help me, right? It's it's showing teachers for the first time, we're going to be able to differentiate instruction for a group of kids that might have a wide range of abilities. You want to teach a lesson on engineering and eighth grade, and most of your kids are reading at a middle school level. Great. But you've got one kid reading at a second grade level and one kid reading at a college level. For a first time ever, we can send an article on that particular engineering standard written at a second grade level, to that one student without embarrassing them. Send another summary of the same topic to the twelfth grade student, written at a college level, and send the rest of the kids an eighth grade version of the summary. And then all the kids can do the activity together, because now the kid who was reading at a second grade level was usually too far behind, was lost and didn't feel comfortable, and the kid who was reading at a twelfth grade level was like, this is too easy, I'm bored, but now you can keep all kids involved and meet them at their level. So it's pretty powerful technology and easy like teachers have done their best for years to try to differentiate instruction, but it's really hard to do as you guys know for sure. I mean, I think there's a real world application for this at the community college level, too. I think that's one of the things that we find is there's a a wide range of academic preparedness when students land on a community college campus in our program specifically. And so this could be really helpful. Graham I'm already thinking about like Summer bridge and our summer bridge curriculum and how we could leverage AI to curate content around that. Um, okay, so let's shift gears just a little bit. You talked about MIT and and they launched a nonprofit. Give us your, your, your spiel for day of AI. What is that? What's the purpose of this nonprofit? How's that? How's that working? So we are a fusion of what I would call young MIT techies and experience read old teachers like myself. And they know the technology and we know how to apply it in schools. And so it's just been a fascinating, small but mighty team. And we really just do three things. Three things. We train teachers, administrators on how to use this technology. We help districts think through policy around artificial intelligence. And we build a curriculum around AI literacy, uh, and for students. And that's what we do, right? So we love to work with kids, we love to work with teachers, and we're in the space to try to help people navigate this new world that we're entering into. My kids like to tease me all the time. They say, tell us again. When there was no internet and the phone was attached to the wall, right? I mean, and that's true. That was the first twenty years of my life. Uh, and now I get to gloat and say, if you guys are ever blessed with children, your children are going to say to you, tell us again when there was no AI. Yeah. Because we're all just going to have to figure it out. Now. we. You guys are much younger than me, but, you know, we had to figure all this stuff out. Cell phones and the internet. And what I'm saying is we just have to do it really thoughtfully when we work with students, whether those are K to twelve students or college students, because even the colleges right now, you know, down the street at Harvard University where I am, um, you know, the the college kids that basically the administration has defaulted to the individual professors to set, uh, the guidance for how to use AI in the particular classroom. So you have some professors fully embracing AI and some professors saying, we don't want you to use it. And so we're going to have to work through this over the next year or two to kind of really come to an understanding about where we're going. How are you helping that conversation? Well, it might surprise you, actually. Um, we don't when I, we work with K to twelve schools, and almost all of them are run by school committees in the country for public schools. And they immediately want to adopt an AI policy. And we say We don't recommend you start at the school committee level, because school committee policies typically run ten or twenty years before they're updated, because of the speed of this technology. You probably have to update you so quickly. Instead, we recommend you start at the high school level. Maybe with your student handbook and start from a state. Of great that a kid needs to disclose. If they're going to use an AI, bot them to use the AI bot or not when they're doing the work, and just begin the process. I talked about earlier dipping your toe in the water. That's what we're trying to do this. Intelligence for the first few years until things shake out as we're all holding on loosely and not letting go. We talked a little bit about the potential, and I certainly want to end later here on the high notes. But but you talked about at the beginning, uh, there are some AI concerns. So what are some of those things that give you a little bit of pause or hesitation? And, and then I guess, you know, a follow up to that is, is what can we do to address them? What are you guys working on there. Yeah. So let me give you just a few examples of some of the problems we have. Voice clones, deepfakes, uh, bias, energy consumption, the amount of energy it takes to query ChatGPT or Cloud or Gemini is pretty strong. Uh, you know, my favorite, uh, is hallucinations, where sometimes the technology makes up answers that aren't true. And I'll give you some examples of these. But the most concerning one, I think, is this phenomenon of AI companions, which a lot of kids have. Common sense media, uh, thinks that about seventy percent of high school kids now have an AI companion or its equivalent. Think of those as like AI, imaginary friends. Uh, and that's great. It's nice to have someone to talk to, I guess, but sometimes these AI companions have gone off the rails and have actually told kids that it's okay to hurt them, hurt themselves, or in a few cases, apparently kill themselves. So. Yeah. So. And we just did a parent toolkit, uh, with Common Sense Media based on the fact that, you know, it looks like seventy percent of kids are using AI or AI companions, but only thirty percent of parents know their kids are doing any kind of AI at all. Right? There's a wide gulf between what kids are doing and what parents know they're doing. And so we're really, you know, trying to get the word out there and get the conversations out there. Uh, but let me give you go back. That's just the most extreme example, hallucinations. Um, at one point, if you asked ChatGPT or or maybe it was Claude where, uh, cheeseburgers come from, it would tell you they come from cheeseburger trees. Uh, and then it doubles down and says, which are found in cheeseburger orchards. Now, I've been traveling the country. Uh, I haven't been to Tennessee recently, but I'm hoping to drive through because I'm assuming that you guys have the cheeseburger trees in your state. In which case I'll be moving there soon. It's possible. Right. Uh ChatGPT. Just fixed about six months ago. You used to ask it how many hours are in the word strawberry? It used to tell you either two or four. Well, without testing you guys, I will tell you this. Three hours in strawberry. And so sometimes the information that's coming from, uh, the technology is incorrect. And so it's imperative that we teach kids to be healthy skeptics when they use this new phenomena. Yeah. And so people are just putting, asking these questions, putting this information, and they're just blindly trusted it. I mean, Ben and I were just talking this morning about one of the the major news stories that's going on today. And, and we were talking about as the story was breaking, like, it's hard to decipher who's who's providing real information and fake information, and it feels like it's only getting more confusing every day. Well, Graham, your point is an incredibly important one. Um, it's a situation now where, I mean, I have chat on my phone. Um, and so I probably use it half a dozen times a day, but at least twice a day I fight with ChatGPT because I know it is giving me wrong information. But the only way you can know that is if you have a knowledge base to begin with. So before we even let the kids have that artificial intelligence, I'm saying. And all kids need to have a sufficient knowledge base. They need to read, write, do math. Uh, be able to think critically, um, because this is a very powerful tool, but it also can go wrong. And so sometimes, you know, like I have two children, I didn't just throw them the car keys when they turned sixteen. We put them in driver's education. And then I had to white knuckle in the front seat with them for several months before they took the test. Right. AI should have a test where kids prove that they have a knowledge base. Because the danger here is if people blindly believe what they get from an artificial intelligence, but they may be getting wrong or misinformation. And if we just completely offload our brain power and we could turn into drones while our brains atrophy or atrophy, whatever that word is and, uh, you know, like, we've got to make sure that kids are thinking first before they can actually use this technology. Well. And when you're using the technology, it's not uncommon for me to ask it to cite its sources. Right. And then look those up separately on the internet. Right. Like you just have to be very careful because, you know, there's bad actors out there in the world and there's bad information being put out there. You know, if you're building a large language model and you're scraping Twitter and Reddit, for your information. Well, I will tell you that things on Twitter and Reddit aren't always the most reasonable or honest or factual. So we've got to be very we've got to teach kids how to use this in a skeptical way. Hmm. That's a great point. I like the idea of of training students on this. You're talking about developing a curriculum, um, around AI, and I'm assuming some of this is is related to the ethics of it and utilizing some discernment when, when and where is appropriate to use this sort of tool. I think this could be really helpful for our students. Um, we are definitely exploring how to expose students to this. How can we allow this to be a benefit to them? Um, in college? And then really, I think there's a genuine, um, value in this for like, career exploration and, and leveraging it in a way to broaden their horizon about what this college degree could lead to. And these tools that are already really established, like Copilot or ChatGPT, probably have enough information to really influence this, this curiosity for students. And so I would be really interested to learn about this curriculum and how that, um, and what you what where you're, where do you start and how do you get students up to speed in a way that's accurate as opposed to, you know, just handing them the car keys and telling them to go drive down I-65, right. Take them to a parking lot first. I mean, just to address your big picture idea, which is spot on. Um, people are worried that, um, AI is going to replace humans for jobs. And what I say is, I don't think that AI is going to replace humans. I think particularly in education, we learned, if nothing during Covid that kids need their teachers and need to be with their peers. Um, I do think sectors will be impacted and different jobs will be created while some of those things happen. But really, what I think is people with AI skills in the future are going to replace people without AI skills in the future. And if that's the case, it's imperative that our students have these skills so that they can be able to function in this new world economy. And I will tell you, there's a bunch of community colleges that we're working with. And I don't know if I'm at liberty to say who they are right now, but are thinking that they're going to start, um, having AI majors for kids because they recognize that community colleges now could be the place where kids go to get real jobs. And, you know, some other kid may go to a four year college and get a degree and I don't know, history or something and may not be as marketable and might have to be a barista at Starbucks. But a kid coming out of a community college with AI technical skills might be highly in demand, right? And so I think what we've seen across the country in K to twelve, anyway, is parents start recognizing the value of vocational education in a way they haven't in a long time. And I think related to that is AI education. I think they're very analogous and similar in many ways. And so what we do is we provide, you know kind of free K to twelve curriculum. All you have to do is go to our website and type in your email address. Don't worry, we won't sell your data and we won't spam you. We will send you a few updates every once in a while. But and you can get behind the curtain and just look at all of our curriculum. You can see our parent toolkit that we built with Common Sense Media. Uh, we're a nonprofit. We're not here to make money. We're here to make sure that our kids have what they need. My biggest fear is our great country created artificial intelligence. Um, and in twenty twenty nine, the Pisa International Assessment, which compares our country to the other big countries in the world, is, for the first time going to offer a discreet AI literacy test for all the students. And it's going to be embarrassing if the country that created AI is ranked number twenty nine on AI literacy, right? We want to make sure our kids have access to this, because I do think it's going to, uh, have a profound effect on their job prospects in the future. Yeah, I totally agree this is aligning right with the conversation. So we had, uh, Doctor Stephen Stryker on who's the director of the Oak Ridge National Lab here in Tennessee. And we were asking him we really brought him on to talk about what sort of workforce needs they have at the lab and the Oak Ridge National Lab. I'm sure it's this way around the country, but it influences much more than just behind the gates. I mean, the entire region is influenced by what that lab brings to that area, you know, and a lot of times it is in these vocational training areas that you've already talked about. But what he told us is that he needs people that are curious and willing to learn. That is the that is the main value he's looking for. He's like people that are willing to learn, have a genuine curiosity, and we ask them about AI. And he said, it's people that are willing to accept that this is going to be part of our workforce, and the workplace are going to advance faster than those who don't. And it's really speaking to what you're saying. Well, and let me say something controversial now, to have kids coming out of schools and colleges willing and ready to learn and still curious. We've got to do a better job in education, right? If you look at the national statistics, kids are reporting being disengaged at school, right. And my hope, my thesis is I hope that we can take a lot of the minutia off teacher's plates and give them the time and space they need to do what they wanted. Got into this profession to do build great, exciting lessons for kids, right? Because we've handcuffed our teachers, and if we do that and we have school, be more engaging, more fun, I think we will build those kind of curious learners, open minded people that we want. Uh, the only time that the student, the national student data shows kids sort of being engaged in elementary school, it drops off in middle school and then is way down in high school. And the only place that data is not true is with vocational schools where kids can see the connection to the real world. Right. And we've we see that. I think we ask our staff, yeah, we've got to get back to having kids be connected to the real world. Learn through play. Understand the different vocations that are out there. Do early college and and be excited about school again. And I hope AI can get us there. Yeah, the completion rates for technical colleges in Tennessee, it's nearly double than that of the community colleges. We actually just got some data from the state today about that. And so your completion rates are significantly higher at your vocational schools in Tennessee versus non vocational. And I think it directly aligns with that engagement piece that you're talking about. So that's that's a great that's that's really interesting. I think there's one person in my life that's really going to care about what I'm what you're telling us right now. And it's my wife who's an assistant principal at a middle school. And, um, this is she comes home and this is a conversation we have a lot. So she'll be really interested to hear about my conversation at work today. Yeah. And look, I mean, I when I, when I speak to teachers, you know, most teachers are parents, right. And I say, um, all right. You're how many of you are parents? Maybe seventy percent of the room raises their hand. I said, okay, let's do a role play. Your kid comes home from school. You say to them how a school? They say good or fine as the traditional answer you get from a kid. What did you learn today? I don't know nothing or I don't know. Those are the two things you get. Now some of that is teenage angst, right? We totally get that. But some of that is because we've overwhelmed our teachers and not given them the time and space that they need to focus on their craft. And I think when I show teachers what this technology can do for them and give them that time and space back, it could usher in a new revolution in education. That's what I'm really excited about. You know, having our kids maximize their abilities, be excited about school and be those open, curious people that, um, the doctor from Oak Ridge wants to have to employ, I can tell how excited you are about the potential that AI has for our teachers and in the classroom, but for the rest of us, what? What should we be the most excited about? What are you like? What what? What do you see as just could be a huge benefit for college students, high school students, teachers, but everyday AI users. What what makes you, uh, think like, wow, this really has some great potential. I mean, there's it's it's permeating everything we do. Right? And what I would say is it's going to if done right and done thoughtfully, um, it's going to help us go further and faster than we ever could. And a lot of people talk about the calculator. And, you know, at first people thought, oh, the kids won't be able to do math anymore. But in fact, kids can do deeper math with a graphing calculator. Uh, the better analogy I like is the moving walkway in an airport. Like, you can get on that thing and it can take you across the airport terminal pretty quickly. But if you start walking on it too, you go pretty fast. It's almost like you're running when you're walking on the moving walkway. And this is what this technology can do for us. And it's going to allow us to to go deeper on things that we're interested in. So the things that students or we are as adults are interested in, we're going to be able to get a lot more information and understand things in a way we never could before. I am not embarrassed to say to you, I you know, I have advanced degrees. I'm a relatively smart person, but I sometimes ask, Gemini, speak to me like I'm a fifth grader when you explain this concept. And if I still don't get it, I say, now break it down like I'm a third grader. And it can do it like that, right? And it's just like, if I had had this when I was younger, oh my God, I would have actually probably done more in different things. Like I just, I needed that personalization that I didn't have. Right. I, I tell people now, I could have been a doctor or a surgeon if I could have had somebody explain this to me in simpler terms, right? But that's not exactly how things work back in the day. No. That's true. My, my. I'm just such a genuinely curious person. I'm obsessed with it. Oh. I'm sorry. I think my internet froze. No, no. I'm sorry. It's important. Oh, yeah? Well, what I would say is, like, if you guys. I don't know if you guys are using it now or have it on your phones, but you can just find out things right away. You can, like, somebody says something on television, you're like, is that true? Or. This reminds me of something. What was that thing? And they can find it for you. And I just think, um, it's a way to engage, uh, and get go deeper on things that you couldn't in the past. Mhm. I'm a prolific Wikipedia user, like I read on Wikipedia constantly. And so this has taken my curiosity to a whole nother level. So you can just go so much further. Well it really can. Right. And I, I for me it wasn't Wikipedia. It was just you know, I used to Google everything. Um, and now this has replaced Google for me. It's just because, you know, you can ask it anything at any time and. Yeah, yeah. Cool. Well, Jeff, I really appreciate you for taking the time to talk to us today. I think this conversation is such a great extension of some of the ones that we've already had about AI, and I think you're providing a ton of insight. I think what you've said, really, for a lot of our listeners, will provide a lot of ease in that this is something that's being thoughtfully talked about. Um, and I'm really excited to leverage, uh, the information that you guys at day of AI have already curated and how we can leverage that with our population of students and our staff. Um, we're anticipating some growth on our staff. And, um, as we do that, I think AI is going to have to be something that we leverage internally, uh, to really speed up, to use your term, um, the way that we do things and I, I really am curious to see how we are able to do that. And I think, um, what you've shared with us today is definitely inspired me to really think bigger about how we can use that at Tennessee Achieves. Well, thanks so much for having me, guys. It's been great. Yeah, absolutely. Thank you so much. Thanks for being here.