Khushali Narechania: K-12 education in the AI era

Hey, I’m Alex. I’m the CEO of aiEDU and this is another early episode of aiEDU Studios. For this conversation, I’m sitting down with Kashali Narania, the director of learning at aiEDU.

So Kashali and I work together. We’re in lots of meetings together. We were just in San Diego together for a conference.

She lives in the East Bay, so we’re even local. But this was really a a rare opportunity for me to sit down with her for a long period of time and and just and just nerd out together. This started out as a conversation about a specific set of curricular products that we had just released elementary explorations but ended up being a much broader wide-ranging conversation including some discussion about what’s coming next.

So I’m very pleased that all of you will get the chance to meet one of the amazing leaders on our team. Kashali is not just super knowledgeable but also just very cool. And yeah, hope you enjoy.

So yeah, let’s get started with why don’t you tell me who are you? Who who is who is Kashali? Well I am in my current role the director of learning here at Aedu.

But I think more broadly I think of myself as an educator first and foremost. I started my career as a high school math teacher and then was in teacher prep at Relay Graduate School of Education for several years leading the content programming there. And beyond that, I’m a wife, mother, hiking and cooking enthusiast as well.

Yeah. Bay Area resident. Bay Area resident.

Yes. Oakland. Kensington.

So, just north of Berkeley. Okay. It’s very, that’s very esoteric.

Most people are just like, "Oh, Silicon Valley." And they don’t realize that that’s actually like two hours away from where you are. Yeah. So when you were in maybe not middle school, but high school, like did you know you wanted to be a teacher?

Yes. There’s some great stories that my parents can share of me in like second grade playing teacher in school. And so that was very clear from an early age that I wanted to be in teaching and in education in some way.

And I think what I wanted to teach, where I wanted to teach, that all evolved as I got older, but I was always in the mentality of wanting to be a part of people’s learning journeys. Yeah. So I mean I think it’s interesting working with so many educators there’s there is a sort of interesting common thread of people who just were drawn to it even though it’s a really tough job.

Yeah. Doesn’t pay well actually in many cases requires quite a lot of education. Yeah.

What was the driving force? Was it just the that that idea of driving knowledge to this next generation or just like being around young people? Yeah.

I think it’s like I’ve always had a curiosity about lots of things. So, I’ve always liked learning myself and like figuring things out and having those light bulb moments are just like one of my favorite things to experience myself. And then when I started being able to kind of experience that with other people and realizing like you can be a part of other people having these light bulb moments by having experiences that they can have where they figure something out for the first time.

It’s it’s kind of incredible. And and I think as I got older also just realizing what what a joy it is to see in like young people’s faces like them figuring out the world around them and getting to be a part of that process. And making it unique to them is is pretty magical.

And so that I think what is what it is. It’s being able to make sense of the world with kids is pretty pretty awesome. Yeah.

Making well making sense of the world also something that’s much more difficult seemingly these days. Yes. Way more complicated.

Yeah. And I I suppose the job of teaching has gotten more Well, would you would you say that? I mean, it seems like it’s harder than ever.

And there’s obviously issues with teacher retention and burnout. You know, you you sort of left the classroom. When year did you leave the classroom to go to Was that 2011?

2012. 2012 Is what? So, it was almost like that was kind of right at the cusp.

I mean, Facebook had been I mean, I remember when I first got my Facebook account when it sort of opened up and it was 2007. Yep. So, I guess 2011 was sort of the maturation of the the new media like social media and you know, sort of like this digital age where, you know, online bullying probably wasn’t really a thing prior to 2007.

Yeah. And definitely not in the scope it is now or like the rate at which it can happen now. I think Yeah.

I think teaching has become harder in so many ways because there are actually more demands of teachers both from state or district requirements and policies. And so there’s a lot more than just teaching really that happens in the classroom. And I think that’s often overlooked in the role.

People often think it’s you’re in the classroom, you’re teaching kids, that’s the end. There’s so much beyond that and in what what teachers have to do. And then I think like the environment that kids are in is changed a lot and it is there’s so much that’s online that used to be physical and trying to kind of navigate that with kids is really really challenging.

So yeah I think it’s gotten a lot harder there’s often times fewer resources allocated towards teachers as well. And that can make it more challenging and we are figuring out new technology as adults that our kids are kind of just immersed in from the beginning and I think that that is not necessarily new but the rate at which it’s changing for adults too is pretty rapid. So we’re in a state of learning alongside our our students in in a different way now.

So this is interesting this this phenomena of students experiencing technology in parallel and in many cases even before teachers. Yeah. I mean I can remember when that was actually not the case like when I was in middle school we had computer lab before we had a computer at home.

We had internet access at school. I mean we we had dialup but it wasn’t very good. Yeah.

And so there was a period of time in in our memories where adults actually could be the arbiters of technology and we could thoughtfully sort of roll it out and you had these structured environments computer lab where students got to learn how to use the internet and and the internet was actually quite a scary place. It was a lot less guardrails and filters. I mean maybe to a to a fault actually to the extent that we have those today but but yeah things have things changed and it was around the time that you were getting into teaching.

Yeah. And even and obviously since you’ve moved into sort of like training teachers that’s only gotten more and more profound. Yeah.

But I am curious if there’s like any specific stories or memories that come to mind where like, you know, a student kind of was showing you something about technology or you felt like you were like a few steps behind your students, whether that’s like the internet or social media or something else. Yeah, I think it’s I feel like when I was still in the classroom, I had a handle on most of the things that were kind of in my students sphere with technology. It was actually when I was working with graduate students and we were doing a lot of like filming.

They were filming their classrooms and getting feedback on their teaching and whatnot. That was a moment where I was like, I don’t actually understand a lot of what is possible u by way of how you record, how you get transcriptions of of recordings, the ways in which you can edit. I had like very, you know, you need a video camera to do this and that whatnot.

It’s like, no, you could probably do it on your phone or there’s another device you can get and here are all these tools and apps that you can use to kind of support with building the clips that you might want for video. And so I think that that’s where I started having these moments of oh wait a second. I actually have I mean like blind spots I think of like what is possible just because I’ve been doing things a certain way or have like a schema that kind of supports one way of doing things with the technology.

And that was, I think, just a great learning moment of like, what else could we do? How else could we leverage these tools that are available online that actually help us better share our practice with one another across space? So we don’t actually have to be in physical classrooms to see teachers teaching, but we can watch them from these videos and give give feedback pretty quickly.

And so that what like that really struck me and I think it just really helped me be able to see like I I have learning to do and I should be listening to what my students or the people who are doing this work like regularly are kind of experiencing to understand what they might be interacting with and how I could be using it as well. Yeah. I mean, even even in the process of setting up, you know, getting all the equipment for for aiEDU Studios, I guess, is the the placeholder name that we’re using for whatever this project is.

Is it is it a show? I’m not sure if it’s a podcast. I guess we’ll see.

But I mean, in the, you know, cuz I used to shoot professional photography and did some video and that was like about seven years ago. And I mean I I I saw the outcome of the outputs of the improving technology and the like the decreasing sort of barriers to access. Yeah.

Just through YouTube. The YouTube has obviously flourished and sort of the creator economy. But I didn’t realize it like how easy it was to get I mean even just listening to some of like the early recordings on these microphones, it’s like it sounds like radio quality.

Yeah. And these mics have been around, but the it was really complicated to interface it into your computer and like there weren’t that many. And even frankly, there just like the tutorials, you know, it was a lot of like written tutorials, like very sort of arcane.

And now there are millions, maybe not millions, tens of thousands of people on YouTube walking you through step by step exactly how to do this. And so it was it was kind of eye opening just how how easy and now I understand like sort of why this creator economy is such a big deal because actually you know now there’s like the only thing standing in the way of somebody creating I think really high quality content is just the content itself you know whereas before and I think even now like a lot of the the podcasts and and like some of the YouTube channels that I see it’s like they’re part of the reason they’re falling behind is because they just they’re they’re still in sort of the mindset of like oh like you know I can just use sort of like my webcam. But there are sort of increasing expectations that go along with these technologies.

And so, and we’re seeing that in to bring us to to artificial intelligence, I think that’s one of the big question marks is as AI unlocks more and more capabilities, let’s say, will or should that actually change the expectations that we have of students? Yeah. Because the internet certainly has.

I think like in many cases, you know, having one or two sources from the library was probably enough for a research project back in the day. And now I think teachers rightfully are like you should have, you know, 10 plus sources because it should take you, you know, 10 minutes to go and find those things. Yeah.

And which brings us to like what what do we need to ensure is in place so students have like understand what’s possible, right? From the technology there’s I think so much of it is just clarity both from teachers and students on like how you can use these tools to create or supplement the work that you are doing and it’s and and not replace the thinking that needs to happen right for students and so I think similar to when the internet became more readily available and you could, you know, search for anything, get a bunch of information and then you ended up with a lot of copy and pasted responses to things. There had to be some work done to kind of help figure out like how do you actually use these sources.

There isn’t it it isn’t just a replacement for your own writing necessarily, but you can use it to help you kind of think through a lot of different topics together. And I think with a artificial intelligence with AI I think similar learnings need to happen of where do we use it? How do we use it?

How do we actually use it to to enhance the work that we’re doing or to help you get unstuck but not completely replace the work that you’re doing? And I think that means like teachers and across all levels are are probably going to have to rethink what kind of assignments they’re having students do and the purpose behind them and being really explicit about that with students and being able to say if you put this into a chatbot, you’re probably going to get this kind of response back. That’s not what we’re looking for.

What we’re actually looking for is some deeper thinking and like what you are kind of bringing out from this. And so, and I, and I, I don’t think that’s a bad thing. Like, there’s a lot of information that my kids will ask me about.

I’m like, I don’t know the answer, but I can look it up and then we can have a discussion, a really interesting discussion about all the things that we find when we look up. But I don’t need to necessarily retain all that information in my brain anymore. Because we have tools that’ll help us kind of access that and then have more interesting discussions about what we find.

And so that’s what I see kind of being really important in the next few years is how do we really get schools and teachers comfortable thinking about where to be using these tools, if to be using these tools, and making it really explicit to students so that they can start developing those skills around how to use it, how to make it something that really enhances their work instead of replacing their work because it’s not going to. So, we’re going to get to that cheating. Mhm.

Top of everybody’s mind and I think for good reason. And I actually I have this whole diet tribe and I was just doing a presentation for a bunch of VCs and I was like people are too quick to dismiss this question of cheating as like just oh it’s you know it’s just something that lites are worried about and they need to everybody needs to get with the program and sort of move past this question of cheating. I think it’s actually for for the educators that I talked to and I know that the ones that you’ve been working with at AID, I mean it is very legitimate.

You know, they they are they are on to something. Their spidey senses are tingling for good reason. And I I’m I’m remiss at sort of how quick some folks are to dismiss those teachers as just not being technology techn technology literate.

And in some cases, they’re not, right? In some cases, teachers don’t really know what to make of language models. GBT etc. but they are experts in learning.

They’re experts in you know knowledge transfer and I think there’s something quite profound about you know there are some legitimate alarm bells. You know I think there the idea that it’s so easy to get answers to questions now. Okay so one of the a fun a fun game that I’ve been playing recently is if I’m at a party and you like you’ll be having a conversation with somebody and someone will ask a question.

You know what? I’m trying to think of the example specifically. It was I think we were talking about the steam engine and the loom and we were trying to decide like oh what was the what was the start of the first industrial revolution?

Was it the steam engine or the loom or you know something else? The cotton gin. And we couldn’t quite get the answer.

And someone was like oh I’ll just like pull it up. And I was like no no no we should actually let’s do this old school. Let’s go and like try to find somebody at this party who might know the answer.

And and it was actually really hard because nobody could remember, you know, their social studies class. But so there’s there’s power in that. But there’s also actually there was a lot of I you know I found myself flexing some interesting muscles not just with myself but with the folks that I was sort of talking to as we were trying to explore and find the answer to these questions and people were sort of recalling different you know books that they have read.

There it was it was a really vibrant experience and even though we maybe didn’t quite get to the final right answer, we were also generating a lot of interesting ideation and also recalling other yeah facts. And even though we didn’t get to the final answer, the the sum total of my learning from that was actually greater than going into Google or chat GBT and just almost immediately getting the answer and then it’s like okay done. And it almost it feels like almost like a real Yeah.

Where you’re getting some cool fact, but then it kind of just immediately floats into the ether. Yeah. So, okay.

So, we have a lot to talk about. I want to start with why you’re here. Besides being super smart and having, I think, some good insights about this topic of cheating that I want to talk about a bit more.

Your team at AID recently launched a new sort of suite of curricular products right before Christmas. We sort of buried the lead there. Yeah.

And put it all out and then like you know feast out for for the holidays. Yeah, just tell us about like what is it that that y’all put out and and just if you can give us kind of like a rough overview of what folks can because it’s on the website now, right? Yeah, it’s on the website.

So there were a few things that we released. First was kind of a re a revised version of our AI snapshots which are intended to be kind of 5 to 10 minute activities that you can do with high school middle school students that are meant to kind of engage students in dialogue around topics related to AI literacy and readiness. For example, one of the really popular ones is about using AI to generate book cover art and kind of who has ownership of that art then and discussion to ensue.

And so these are there we have about 42 of those. Those are for by content area for math, ELA, social studies, science and a couple general ones. And these have been really wellreceived because they are very easy to kind of implement without a lot of a lot of extra kind of internalization that needs to happen and kind of fit right in with warm-ups, with closing activities as a good transition between between different kind of learning experiences as well.

So, that’s up and running and I think have been really, really, really well received by all the folks that have tried them out for us as well. So that’s one big one. The other big pro, product that we released were projects specifically geared to the AP computer science classroom.

And so these are aligned to the College Board standards for APA computer science principles and explore again ethical ideas in AI readiness and literacy and range from you know a couple hours of instruction to up to I think six hours for students to produce at the end of some sort of either presentation or conversation on these on these topics as And along with that, we have a smart teach manual, which is kind of quick ideas for how to think about integrating AI literacy and readiness to an AP computer science principles classroom as well. And so those are kind of the big things that have come out. And prior to all of that, there was also the readiness framework which kind of is helping us anchor all of the content that we’re building on kind of naming some of the key skills and concepts that we would want students to kind of master and have in their have in their arsenal as they graduate from from K-12 schools and into into career.

Yeah. I I feel like you’re giving short thrift to the the readiness framework because it’s I think it’s actually big deal. Yeah.

The you It’s interesting because snapshots were that was like one of the first products we released. What you inherited when you joined AID was a much bigger corpus of snapshots. I think it was like 180 Yes.

Snapshots which was way too many. But I think it’s in in sort of in keeping with the approach that we had early on was we can’t go to teachers and say here is sort of a new curriculum that’s AI. In fact, we actually tried that.

We built a class. Was like the quarterlong I think maybe even a semester long course. Yeah, 10 week 10e course called introduction to AI and you know we found is unsurprisingly there were very few teachers that had the flexibility to say oh I’m just going to completely scrap everything that I’m doing and and plug in an entirely new course.

So we went almost to the opposite end of the spectrum with these sort of like very bite-size easy to integrate sort of like between classes sort of like as bell ringers. And I think what we found, and I’m curious if you have any examples of this with our like professional learning communities or or like fellowships, you know, while they were designed to be five-minute activities, we I feel like there’s a lot of examples where teachers would say, "Oh, yeah, you know, this started out as a fiveminute classroom discussion and it was the students were so engaged that we ended up creating an entire sort of module around it or it extended into an entire class period." And I think but yeah, can you just tell me more about like is that something that you implicitly are expecting teachers to do or is that more sort of like a happy accident? It’s I mean I I think they were one of the things in this new release of them is we explicitly call out that they take around 10 minutes, not just 5 minutes.

I think some are doable in 5 minutes, but realistically to have students engage in conversation and and kind of really think critically about the questions being asked, you need more time for that. So so I think we’ve more explicitly named that these should be deeper conversations. And you can do it in 10 minutes.

And we’ve had our fellows who’ve kind of tested out these new the new snapshots. We’ve watched them do it in 10 minutes and we’ve watched some that take 20 because they’ve actually integrated it into their lesson and have brought in some of their content to the snapshot and and like kind of built out the discussion a bit as well. And so our there’s two ways to go with it.

And I think we want as many people as possible using these. And so the easiest entry is say we have everything for you to facilitate it in 10 minutes and here’s what it can look like. And then if you’re interested in doing more, that kind of becomes unique to what your class is, who your students are.

And so there’s all these other ways that you can think about having a multi-day discussion or building it out. And there is like appetite for that and I think there’s like great like community amongst the teachers who have been using this to kind of share how they’ve been using it as well to to to personalize for their classrooms. So I think that’s that’s like an exciting new place to look as we kind of build out more resources of how can this look in different formats beyond the 10-minute warm-up.

Yeah. Yeah. The I mean I think it’s a testament to we are not you know our goal our vision is never it has never been to be the sole arbiter of AI curriculum or for there to be some sort of like reference curriculum that like all AI literacy is accomplished through the AIED curriculum package.

But I think you know the so one of our one of the early school districts that we engaged with just to learn as as part of our learning process was Gwinnet County Public Schools and they were probably one of the first districts in the country to roll out like AI literacy and the flavor of how we think about it which is all students having access not just sort of something that gets slotted into you know STEM or computer science pathways. And they have this sort of I think very clever you know way of thinking about this which is the the sink the swim snorkel scuba Yeah. Metaphor and it’s like everybody should learn how to swim.

Yeah. And then some students will will want to snorkel and then a smaller group will actually go scuba diving. And I feel like snapshots is almost like we’re going to sort of a desert where there’s there aren’t even like waiting pools.

Yeah. Yeah. And so the first thing is like if you want students to swim like they have to have a place where they can actually get in the water.

And I think right now teachers are sort of looking around. They’re saying like I don’t even know where how to get their feet wet. Let alone how to teach them you know the freestyle and like put you know the snorkel gear on.

I’m belaboring the analogy perhaps a bit too much. But I think there’s you know once you build the pools and again I’m going to really belver this analogy. Once you build the pools, you know, we don’t necessarily have to go and show everybody exactly every single stroke, there are going to be some teachers that are going to take take that as sort of a blank canvas and run with it.

But I think, you know, until you can actually sort of lead them to the water and sort of help them get their feet wet. It’s just like really it’s unclear where to start. Yeah, I think that’s right.

And I think that’s part of what we wanted to do in these updated materials was like build more for teachers to have like building their own confidence and kind of facilitating these discussions. And so there’s a lot of notes for the teacher on like here are some possible misconceptions or here are the ideas that you might want to kind of hone in on which I think is intended to help spark for them like here’s what I might want to know more about and being able to create more of those spaces and comfort for themselves and being able to lead the discussions cuz I do think a lot of times teachers will potentially shy away from engaging in the topic because they themselves want to learn more or want to have kind of more of an understanding. And so, our our desire in kind of upgrading these materials was to build in some of that like learning within the material so they can have that as they’re trying to facilitate with their students as well.

And so it’s I think they serve as a great space for both the teachers and students to do kind of that swimming in the AI readiness content. Yeah. Because you did some one of the first things you did besides sort of overhaul the in fact it I think it was a pre precursor to overhauling the content was doing some research with teachers and so is that something that came up this sort of like this sort of confidence sort of roadblock where they were just weren’t feeling like they had the mastery and that and teachers do feel like they’re supposed to master a subject and that’s generally the model is like they become experts in whatever it is they’re teaching and then they kind of sort of structure the class.

This is a very tough subject for teachers to become m you know gain mastery over. Yeah, it’s it’s so new. We are still learning so much about what LLM technology is, how it applies, how it actually works.

And that was like very clear in our research actually. It was one of the big headlines to come out of it that like educators have a lot of uncertainty and like into fear in some ways around what this technology is and how they should be using it or addressing it or learning about it. And and I think our one of the things that we can do is help disseminate more information that is digestible and appropriate for teachers and students so that they can build their knowledge and then I think very much do the thing that we’re we’re saying is important is like think about this critically and decide on when to use it, how to use it, what to do with your students cuz at the end of the day teachers are the ones who know their students the best and so they are the ones that should be making the decisions about how to implement this any of these curricular resources for for for their students.

And we just want them to have everything to help them feel confident about how they’re implementing it. And that was again driven largely by what we heard from teachers that they don’t know enough and they want to learn more. And so we can we can help provide some of that and some guidance on that.

Yeah. So like I keep beating this analogy. So, so you you build the pools and you maybe it’s not just a a waiting pool, but there’s like you can go a little bit deeper.

It’s not necessarily like a diving chamber yet. Yeah. But also you need to build the roads to the pools.

Yeah. To have locker room. I mean like you need like you can’t just put something out there.

And I think that was one of the things that we saw, you know, really early on is there’s actually there there are curricula that address artificial intelligence. And certainly if you are looking for self-guided learning and YouTube is maybe the best and worst example because there’s just like so much and it’s just a ton of noise. But how do you navigate that?

What is actually quality versus not quality? I think like there’s a sometimes like you can actually get overwhelmed with the amount of information. I think that’s the issue with sort of like where we are today where there’s certainly not a lack of discussion about AI.

So like or a lack of awareness at this point. And that’s actually changed our you know because when we were you know before your time at AIEU and before Chad GBT you know AI literacy for us was really introducing the topic to people for the very first time and for most people they hadn’t even connected the dots with like oh I’m using you know AI in my phone when I use Google maps or AI is making content recommendations to me and so there was some lowhanging fruit there of just sort of like trying to build comfort level with this is not actually science fiction anymore this is something that’s sort of quietly creeping into our lives. But now we have chat GBT and every education conference is an AI conference now.

Yeah. So it’s not so much a lack of awareness, but there’s still I think teachers are they they they don’t know really where to tack this problem because for most teachers, which brings us back to this question that you kind of led with, which is what do I do about how should I be using chat GBT? That’s actually their first instinct is it’s not so much like how do I teach students about AI more broadly but how do I use this and then how do I deal with my students using it.

Yeah. And so this is a way of of talking about the framework because I think when we were going about thinking about okay what is the new curriculum we want to build I remember having these discussions and there was a big sort of fork in the road I think for us and even thinking back to some of our early conversations there was this question of you know should we be basically teaching students to prompt engineer right and to become power users of AI technology which is different from the framework that your team published because the framework actually it addresses is AI literacy and as one of three buckets. Can you do you remember the can you can you just spout the three off?

It’s it’s AI literacy is the first domain. It’s like know the basics the basics. So it’s how to use the AI and then being a critical thinker and is the next one and then the third is knowing the kind of human advantage going into it.

Yeah. So, so mastering chat GBT fits I guess certainly the first domain and then also the second I think you can argue that prompt engineering in the abstract is actually asking critical questions or knowing how to ask the right questions. And then also sort of like critically evaluate the outputs.

But it is it is a far cry from what I think other other folks maybe expected from us. But I it brings me to sort of this question of like how to deal with cheating. Because there is there’s a school of thought that like the way you deal with cheating with students using chatbt is it’s a technical challenge.

We need to come up with policies and academic integrity procedures and we need to identify the perfect detector tools, right? And we need to design foolproof assignments that can’t where students can’t use chat GBT to cheat. And it’s actually which is a very different question from I think the way that the framework really addresses this, which is what should your what should your learning be oriented to?

Like what are the outcomes that we want from students to be able to to have the knowledge they need to succeed in a world where they are going to be able to go home and use these tools to generate content and answer questions and write essays. Yeah. Yeah.

And it and it seems like what you’ve you the the the place where you’ve planted the flag is durable skills. Yes. Communication and collaboration and you know creative problem solving self- agency.

Yeah. And then so the answer to how do I deal with cheating is more like well how do you adapt your assignments to focus on building those durable skills? It’s it’s and it’s not and I hesitate to say like this doesn’t all fall on the teacher changing their entire curriculum because I think that is a really daunting thing to hear that I have to rethink everything that I’m having students do.

But it is really about I think like a great counteram or not a counter example another example of this in a similar kind of vein is when calculator technology became readily available and we went from a time when like doing any sort of computation was very like labor intensive and there was no quick way to do it and so a lot of math revolved around understanding how to do those computations and arithmetic Here come calculators, especially graphing calculators that have really significant capabilities. And you go from a situation where you’re so focused on arithmetic to now you have a device that can do a lot of the arithmetic for you so that you can do more complex problems, right? And the I think the education community really grappled with like what do we do?

Are we banning calculators? Do we focus only on arithmetic or how do we integrate it into curriculum so that it’s a part of the learning experience where they actually have to students have to learn how to use the calculator how to use it effectively and what does that mean for what we can actually ask students to be doing and focusing on and I think overall like I think we can all agree are a part of our life we use them students know how to use them we have you know state tests adapted to it there’s calculators are generally allowed on almost all portions of of testing and if they’re not it’s very explicit and there’s you know rationale provided for why and so I think of that and I think we can learn a lot from that like we’ve done this as a as a society where we’ve kind of welcome this technology in that’s kind of opened a lot of access and doors to what we can do and how we focus and so similarly with I think AI tools it’s I think if we try and close the door to it and say this is not allowed and we should not be using it we should just have really strict policies in place. Just inviting people figuring out how to work around that instead of it becoming something that we welcome into learning spaces as as a tool.

And we teach students how to be using it, how to think about using them effectively. And for teachers to then be thinking about like what do we need to do in our coursework? That needs to change potentially that can stay the same.

That leverages this tool just like in math we have a shift like there is a distinction I think you can see in what was taught before calculators and what was taught after calculators and so yeah so I think that’s just like how do we think about what is it that we want students walking away with and how do we use AI as a tool to help us get there and that will take some thinking and I think it’s thinking with teachers and students and communities on what those what those skills need to be and will be most effective for for the set of students that you know any one community is working with as well. Yeah, I might I actually want to I want to I want to push a little bit on on this and it’s it’s almost to sort of pull back the current on an internal I don’t know if I’d call it a debate, but certainly like a vibrant discussion that we’re having within our organization about what does it mean? What does it look like for us to say, okay, teachers should embrace the tools.

Where is it maybe not appropriate or or just not the quite the right moment for that to be the the guidance. Yeah. Because some of the questions that I’m grappling with is if we’re going to go and say, well, we need to build around these tools in the same way like if we use the calculator example, well, if you’re going to make assignments that are implicitly or explicitly requiring a calculator, we have to make sure every student has access to a calculator.

And I think that’s we’re not quite there yet. Yeah. You know, in schools.

And so it the the it’s like that’s a conversation we have to have sooner or later. Yeah. But my worry is that for many schools, they’re actually not even there yet.

Like they they kind of need to sort of operate a few steps earlier. And then there’s also this question of while the calculator I think is an apps analogy for many reasons. You know, there’s something about language models, and I think this is why it’s had this zeitgeist moment, way bigger than the calculator, right?

I mean, the calculator, I think, was a very big moment for educators, but most people outside of education probably didn’t realize or couldn’t even tell you what a TI89 is, right? Everybody is paying attention to Chad GBT. Yeah.

And and that’s because there is something in just in our guts. It’s like when you see it write essays and press releases and marketing memos and poems and songs and emails and and love letters and you know wedding plans and itineraries like there’s there’s a breadth of application. Yes.

That yeah, I just I I I wonder if if we we might need to preserve the experience of not being able to do something because if if we if we if if the learning if the if the learning in school is basically how do I use these tools to basically make everything easy and then if everything becomes easy there is I mean I feel like part of the point of school is also just sort of like the the process of you creative struggle and productive struggle. And it it seems to me that there’s more to education than just how do we make your life as efficient as possible. Oh yeah, 100%.

And I think that’s that goes back to what you’re saying is like what are the outcomes that we’re driving towards with being in school and for students to have, right? And I think it there’s there’s just like a lot of work to be done around continuing to foster curiosity and problem solving. And so I don’t like in no way do we I think do I ever advocate for like schooling should revolve around how to like integrate AI tools in everything that we do.

But it’s actually like the step above that which is how do we think about when we want to use this and why we want to use it and is it aligned to the kind of goals that we have as a school and for our community and for our students. And being really thoughtful about that process before even bringing anything in front of students. And I think it’s really important to actually, especially for our older older grades, older students, like they should be a part of the conversation.

They are experiencing it outside of school. Most definitely. And and and so like how do we actually be really transparent about it with with students and have them partake in the solution making because that is actually the thing that we are wanting to prepare them for.

And here we have an opportunity to really have their voice be centered also in kind of like how we think about solutions around this. Yeah. Which actually that brings us to something that we’re working on which is trying to create like a workshop or a learning experience that either directly or implicitly include sort of the student perspective on how how would you like rather than trying to figure out how do you prevent students from using AI tools help teachers understand how would students use AI tools to get around your assignments.

Yeah. And you know, to me, the first step in addressing cheating is you have to get into the mind of the cheater, right? And and for the most part, they’re not they’re not evil, conniving, you know, mischievous people.

They are, you know, enterprising kids who have soccer practice or band or friends and, you know, like this assignment that where they have to write a book report about someone that they don’t really care about. Yes, they’re not. It feels like they just have to go through the the motions and it doesn’t feel like relevant to what they’re interested in.

And so sure they’re going to it’s and it’s also just I think it’s unfair to you know the idea that like you can literally type in a sentence and press enter and get an entire assignment like so now we are basic it’s like I the the analogy I’ve used once with somebody which maybe resonates with educators is vaping and so someone’s like addicted to nicotine and we’re trying to we’re telling them okay well we we don’t want you to use nicotine but then they are in this world where in front of them is like, you know, a whole matter of, you know, vapes right in front of them. Yeah. And and they’re flavor.

They’re candy flavored. And you know, even very disciplined adults struggle with quitting right nicotine as unfortunately a lot of high school students these days as well. But it strikes me as like it’s really hard to to expect a student to like be able to resist the the tantalization the tantalizing opportunity of like, you know, one sentence, press enter.

Yeah. And so, you know, it’s to me there’s like actually some a sense of almost respect of like this is not a value judgement on, you know, you’re a bad person for trying to get around the assignment, but actually designing an assignment that’s like where they can’t just type one sentence and press enter is almost like it’s it’s helping them get to the right approach to learning without sort of expecting them to sort of be holier than thou, right? One of the one of the things I want to do on the show or podcast or whatever this is is kind of like get into real world experiences of educators.

And so one of the things that I came across I follow a lot of like the educator subreddits. I’m going to paraphrase, but one of the one of the subreddits that I found there was a a marketing professor. So he’s in community college and he was posting he’s he’s figured out his way around to get around chat GBT.

Yeah. And so basically he says, "Oh, you know, I figured out basically, you know, how to get around this and the way I structure my assign one of my assignments is it’s basic it’s I think he has a a lecture about how to write like a marketing a good marketing email." And so he he says, "Well, what I do is I just assume that students are going to use chatbt." And so what I tell them is I want you to use chatbt to write a marketing email. Yeah.

And then I want you to critique it. And and then one extra layer he says and then in the criticism reference like specific slides in my lecture that they have access to y as part of the critique. And I’m curious I have my own opinions but I’m curious from your perspective like is is that an example of a chat GBT proof assignment?

Like has he solved the problem? Yeah. Of I think it’s a it’s a good first attempt.

The the our chat models have increased their ability. So you could actually feed the feed ChatGPT the slides and pose the question. Now go back and critique the response you just gave me and use the the slide deck for some citations.

And you’ve just used ChatGPT to do the critique which is the thing that I think the professor was trying to not have students do. And so I think there like adding the additional layer of critique is I think a is a good direction to go but again you can use ChatGPT cloud to kind of do that critique. So so the the prompt would basically go so first you say, "Okay, write a marketing email about whatever the topic is." Y and then you would copy and paste maybe create a new a new chat and say, "Okay, your your your assignment is you’re supposed to now critique this marketing email and chat GBT allows you to upload materials." And so you upload a PDF file of the the PowerPoint.

Yeah. You say reference this PowerPoint and in your critique site the specific slides as part of your critique. Yes, exactly.

And that would take all of maybe five minutes. I mean I think in total maybe 8 minutes and then let’s just add another 15 minutes for a little bit of massaging to make it look and feel a little bit authentic. And you can even say like oh you know make some grammar mistakes or something so that it sounds like a community college student wrote it.

Yeah. Yeah. So I guess so I agree.

I don’t I don’t think that’s a good I don’t think that’s a solution. I think that’s actually it’s an example of just why it’s so important for educators to just come to peace with the fact that they are always going to be a few steps behind their kids. Like it’s going to be hard for us to outsmart these young enterprising Gen Z and Gen Alpha kids who are steeped and immersed in not just what this technology is but also like you know like how to content is like one of the most sort of common forms of content on on Tik Tok on Instagram.

So so yeah I guess what is could we help this you know well-intentioned college professor what would it look like to to adapt the assignment a little bit so I think I come back to like what’s the purpose of this assignment is the purpose for students to be able to draft a really strong like marketing campaign email is it for them to be able to identify like the key components of a strong marketing campaign or maybe more like is it for them to be able to actually draft something that they then test and see if it’s actually effective with like a with the demographic that they’re trying to reach, right? And so, the last one I mentioned because I think that’s where you can do some interesting things where it’s like have ChatGPT write your campaign. Now you actually have to take it out to like three or four different people.

Have them read it. Did it resonate in the way that you thought it would? And like have an have some sort of dialogue or interaction with other people outside the class or even within the class or bring it to class live.

We’re going to look at it together. Where do we see it meeting the criteria that we’ve talked about in class? Where do we see it not meeting the criteria?

How do you know? And you can do live polls with students to kind of assess it. And so that that’s another way of kind of bringing I think it’s how like and both of those things I think it’s thinking about like what’s the human connection back to this.

How are we bringing other people in? How are we talking about this together instead of it continuing to be an individual assignment that is completed and turned into the professor? Yep.

That’s one way that I would think about kind of doing more to make to make more out of that assignment. Yeah. Okay.

So to let me let me try to abstract this. So one of the things I’m hearing is you know I think teachers educators are increasingly going to have to treat in-class time as one of the most important moments of their their educator educational toolkit. Y and they need to use that time to really try to build the skills that they where they need to just be be sure that students are actually doing it themselves.

And so I feel the same about like writing an essay. Like I don’t think, you know, I often will say like the age of the take-home essay is dead. And I do think that’s the case.

But I don’t think that students don’t no longer need to learn how to write essays. It’s just that I think the take-home essay is dead. That’s probably something that needs to you need to spend more class time having students write.

Yeah. So I like this and so you can imagine, yeah, I guess the kids could come in, they’ve used chat GBT to write, you know, try to write the best marketing email and so and maybe what I would even the one thing I would also do is I would before almost any assignment I would do a share screen and I’m like I’m going to pull up chat GBT and I’m going to show you what I know all of you the first thing you’re all going to do I’m going to copy and paste the assignment plug it into chatgbt and press enter and so like that is the baseline. Obviously, I expect something better than that.

Yeah. And so, it’s up to you first of all to come to class with something that’s clearly better than whatever the bas you know the basic thing that chat GBT is. Yeah.

And and then using that in class time and then for what happens out of class. Yeah. What are the like basically trying to hone in on what are things that just chat GBT can’t do?

It can’t go and have a conversation with with people, right? And and you know what if a student and I think there’s a certain point you get to a place where if a student is really savvy they could have chat GBT create different personas of people and then like have the conversation and like use 11 labs to like you know create the you know like voice like real like synthetic voices that are actually sort of like responding. And you know what?

You might just you might at a certain point say, you know, if a student is going that far, right, I’m I’m either like it’s just it’s going to be a small subset of kids and so I’m not gonna like over pronate towards that. And also maybe there’s actually that student is demonstrating you know a lot of sort of enterpreneurial spirit and skill that I I’m just going to sort of agree to look the other way. But but what you’ve also pointed to is that sort of thinking back to the calculator, what the calculator unlocked is the depth and the complexity of what we can do in a math class increases.

And so maybe the same applies to this marketing class where you know what you can actually go way beyond, you know, the assignments that you had before. Yeah. And so yeah, I think I think that works.

Think it’s easier said than done because I think what often teachers will say is well but there’s other stuff I need to get done in class. Yeah. So there it’s it’s I I never want to and I think this is important is like there is no sort of quick fix right to these questions but I think we can actually at least get directionally on the right path and and certainly identify where you know I think the worst case is that a teacher thinks they have designed a foolproof assignment and they’re blind to the fact that they’ve just you know they’ve just added one more extra step that might not be you know like challenging students the way that they think right and I exactly and I think like the more and more I’ve been in the world outside of like my own school, the more I realized there’s so little that I actually that anybody really actually does in isolation in like a take-home assignment kind of way that we I did so much of in my especially high school experience.

And I think there is something around like how do you build in more of those collaborative experiences which I think reflects what life will be like for ever after school which is you’re going to be working with groups of people and completing assignments like together right projects together. And I know group work is notoriously challenging to do with with kids but also I think some of the most meaningful work and like skill building for for kids as well. And I think that that there is a opportunity here to figure out and think about how do you use this new technology to kind of help pull out those skills more of like what does it mean to actually work on something across people and across contents.

And I think that’s that’s really exciting because I think that’s reflective of what most of career and life readiness means as well. Yeah, this is I mean brings us back to durable skills and and the framework. And I think that’s that’s the I that’s the the the silver lining or the copper lining in all of this is the at the end of the day being able to write a good marketing email by itself is actually even if even if you were able even if there was a perfect chat GBT detector and what you were able to do was get around students and force them to go home and just write the email themselves.

The problem is writing a good marketing email is not that that skill is not going to command economic value in the way that it did two years ago. And which brings me to actually this is so so I mentioned my little anecdote about this this nerdy discussion that my friends and I were having about the the cotton gin and and the loom and the beginning of the first industrial revolution. Was because I was basically kind of I was talking about I I just watched this lecture from David Atur at MIT and he was talking about some of the research that they had done in with language models and sort of the way that they’re impacting people in in workplaces that have actually started to implement them think like customer service and one of the insights that he had was it’s it’s not actually the the he was like I’m less concerned about AI like destroying like about jobs disappearing like there will be jobs that disappear there’s going to be more new jobs.

But the the the bigger challenge that we’re going to have to face is that the economic value of certain types of expertise decreases. And so while there might be lots of jobs, they aren’t necessarily going to command the same wages that they did before. Yeah.

Yeah, you know, so you might be able to write more marketing emails and there’s going to be people at companies that need to sort of be at the wheel and using Chad GBT, but the skill of using AI tools is going to be commoditized. Yeah. And so and this and and so this is like I think to me the the the meta takeaway there is what is the hard what are the hardest skills to commoditize?

It’s those durable skills. Yeah. And that was like and Sam Oldman actually he was this was like actually three years ago he was talking about you know the jobs that are really going to disappear people are going to spend less time in front of their computers.

Yeah. And he was was very apt and I think the and that actually sort of marries with one of the ways that I would talk people ask me like well what is my job at risk you know it’s it’s hard to say what specific job is at risk but I think if someone wants to know whether whether they are equipped to succeed or maybe fall behind in the age of AI I think it’s like open up your calendar and look at how many meetings you have. Yeah.

And if if your workday consists mostly of time by yourself sitting in front of a computer, whether it’s writing code or writing advertising copy or press releases, whatever it is, if you’re content generating or doing research, you know, that that should give you some some concern, right? Because I think that’s the type of work that is going to be automated. Yeah.

What’s harder to automate is, you know, having a conversation face to face with a client, a customer, a colleague. Yeah. And and then that brings you to, okay, well, if that’s if that’s where we’re heading where students need to be able to have more of this sort of like do more of the human aspects of work, well, what are the skills they need to have to be able to engage with colleagues or really annoying colleagues or really annoying clients?

Yeah. Because that’s really where this push comes to shove. I mean like a great colleague is a great colleague but some colleagues are we don’t have any difficult colleagues but but obviously many people do and and you know that experience of also just like dealing with like you talked about like group assignments as being some of the hardest things that students deal with but actually even when you fail as part of a group you learn a lot about like why did you fail?

What was your role in the success or failure of that? That. Yeah.

I mean, I think it I think that’s it. And it’s like I was just we were just having a conversation with our 8-year-old about some of the things she was experiencing with like some of her friendships and people that she likes. And we had to explain to her it’s like people aren’t like math.

What you do with one person might warrant one reaction one day and another reaction another day. And it hasn’t like you haven’t changed or done anything differently, but something in their world has. And you have to be aware of that.

And what you do with one person might be different than what you do with another person because they need something else. And I mean with an, you know, there’s so much of that social learning that happens so young, right? But I think that’s what you what we’re trying to instill with, I think, all of our kids, it’s like how do you actually interact and interface with others such that you are able to do the best kind of work you want to do and bring out the best kind of work from others.

And that requires being able to understand and like read cues and not take things personally and being able to like confront head-on when something isn’t going well. And how do you do that effectively? And I think we have a lot to learn from our younger classrooms because they do a lot of that work explicitly cuz it’s so much a part of what younger like what the younger kids are working on in school is just understanding how to interact with one another.

And I think we can bring some more of that back to our upper grades, too. Yeah. Yeah.

That’s why that’s why I I it gives me pause when some of the excitement around AI tutoring gives me pause because I think it it assumes that the most important thing in education is just sort of like the knowledge transfer and like learning the things. But I worry about a world where students are all just sitting in front of their computer. Even if even if like scores go up.

Yeah. There there is there is something to be said about just like especially now that kids are we know they’re going home and they’re interacting with AI companions and we can’t necessarily stop that or teachers can’t right they don’t have any control over it. So so school I think increasingly is going to be the place where we can ensure the one place where we can guarantee that students are like building those those soft skills.

Yeah. And of course the hard part is, you know, it’s easy for us to say that to the teacher who is measured not on soft skills but rather on the hard skills and the standards. Y and I think that kind of brings us back to like the curriculum that we’ve been building.

There’s, you know, we have we we’ve been paying a lot of attention to how do we align this to the standard so that it can check both boxes and and we’re not this is not some crazy insightful thing like there’s a lot of other really high quality curriculum out there that does both. Yeah. And project based learning while we utilize it throughout our curriculum is also not new, right?

Like we didn’t invent the idea of project based learning but this is something that schools have talked about for a long time haven’t necessarily you know mixed it’s mixed results in terms of which schools are doing a really good job implementing it at scale I think they actually often times really struggle so okay to close things off I just want like you’re you’re not done unfortunately like the treadmill sort of continues as a as a curriculum developer what what are you excited about that’s coming next you know I know that you don’t have to without sharing too much but maybe like one thing that you can sort of share with our small but but loyal audience that that they can expect in 2025. Yeah. Well, I’ll share two things because I think that there it’s really exciting.

I think we AID has kind of focused on secondary in the past and I’m excited to kind of bring elementary into the mix. And so we’ll have actually some content specifically for elementary teachers on like how how to kind of bring some of these concepts into the classroom, which like we were just talking about, it’s actually skills that they’re already working on in their ELA and math standards. And and so that I’m really excited about.

The other thing that I’m really looking forward to with having kind of built the AI readiness framework and the competencies for students, we have been mapping kind of what do learning experiences look like across grades or within grades that kind of can build around these competencies. And so, thinking about packages of content, that address the ethical implications of AI for, say, sixth graders, like what are some warm-up activities, some lessons that you might be able to use and projects. So a teacher who is interested in that topic or their students are have kind of a bigger bank of resources that they can kind of pull from to to support kind of a a learning experience across across that arc too.

So more to come on that but really excited for kind of these these resources that are kind of going into younger grades and kind of deeper learning as well. Yeah, I’m I’m very excited for the elementary school stuff. It’s something that people ask for all the time.

And it it’s not that we don’t think it’s important, but it seemed like the the most important place to start is like the future workers who are like, you know, only a few years away from, you know, moving into sort of their post-secary, you know, pathways, whether that’s career or college. But you know, anybody with young kids will tell you like there is a ripe opportunity to begin conversations and actually start to orient them so that we’re not, you know, because a lot of stuff is baked, especially like I know like career exploration, you know, a lot of the research suggests that, you know, the fundamental the most important moments for, you know, people sort of self-identifying with different careers happens like a lot younger than people often expect. Yeah.

Excellent. Well, this was really fun. Thank you for yeah experiment.

This is legitimately the first show that we recorded. I think it went pretty well. Yeah.

So yeah, Kashali, thank you so much for joining us. Director of learning. Thank you so much.

Alex is great as always. And if anybody who’s watching wants to chat with you, follow up how can they get in touch? You can reach me at kosaliedu.org or if you want to reach out to the learning team more broadly, learning aedu.org.

Okay. Amazing. Thank you so much.

Yeah. Thanks.