Your AI Injection

The Power of AI in Project-Based Learning with Bob Lenz

November 16, 2023 Deep Season 3 Episode 8
The Power of AI in Project-Based Learning with Bob Lenz
Your AI Injection
More Info
Your AI Injection
The Power of AI in Project-Based Learning with Bob Lenz
Nov 16, 2023 Season 3 Episode 8

In this episode of "Your AI Injection", Deep interviews Bob Lenz, CEO of PBLWorks, on the transformative power of project-based learning (PBL) in education. The two discuss how PBL engages students more deeply than traditional methods, fostering long-term retention and application of knowledge, particularly in underserved communities. The conversation then delves into the role of AI in enhancing PBL through project design, personalized assessment, and teacher support. Bob shares PBLWorks' strategic plan to use AI and digital platforms to expand the reach and quality of PBL for teachers, aiming to make learning experiences more ubiquitous, effective, and engaging within the next five years.

Learn more about Bob:
and PBLWorks:

Learn more about the applications of AI in education:

Show Notes Transcript

In this episode of "Your AI Injection", Deep interviews Bob Lenz, CEO of PBLWorks, on the transformative power of project-based learning (PBL) in education. The two discuss how PBL engages students more deeply than traditional methods, fostering long-term retention and application of knowledge, particularly in underserved communities. The conversation then delves into the role of AI in enhancing PBL through project design, personalized assessment, and teacher support. Bob shares PBLWorks' strategic plan to use AI and digital platforms to expand the reach and quality of PBL for teachers, aiming to make learning experiences more ubiquitous, effective, and engaging within the next five years.

Learn more about Bob:
and PBLWorks:

Learn more about the applications of AI in education:

Deep: Hi there. I'm Deep Dhillon. Welcome to your AI Injection, the podcast where we discuss state of the art techniques and artificial intelligence with a focus on how these capabilities are used to transform organizations, making them more efficient, impactful, and successful.

Hi there. I'm Deep Dhillon, your host. And today on our show, we'll discuss the potential benefits of AI and project based learning methodologies for improving educational outcomes. I'm super excited today. We've got Bob Lenz, an education leader who's reshaping how students learn, as CEO of PBLworks. He's driving quality based project Learning for all students promoting success in college and careers.

Bob co founded Envision Education that also helped pave a path to college for underserved urban students. Bob, thanks so much for coming on to the show.

Bob: Hey, thanks, Deep. Looking forward to the conversation. 

Deep: Awesome. So maybe we can get started. Can you just share a little bit about your background and maybe talk to us about what, what is project based learning and what are some of the nuances and advantages of project based, uh, learning?

Bob: Well, my background starts in the classroom as a teacher in middle school here in San Francisco and North Beach in Chinatown. Most of my students there were families that immigrated from China.

And I was actually doing project based learning, but I didn't realize it. Part of the reason was I was fortunate enough when I was in the fifth grade to have a teacher named Mr. Cooper who did projects with us. So I actually still have the poetry book that I created and had a poetry reading Mr. Cooper put together when I was 11.

And so when then I was, when I started teaching, I actually did that same project with kids in San Francisco, but we were able to bring in poets from City lights, bookstore, the famous bookstore, Silverstein wasn't, wasn't he on there and city? No, it was, um, one of the beats guys, Lawrence, but, uh, and then I, I moved to, uh, up to Marin County and taught at a high school where we transformed the school, basically using a project based learning.

And we formed career academies for juniors and seniors. We were very successful and got lots of recognition. I got to go to the white house. We're one of 13 new American high schools in the country in 1999, um, when, when Bill Clinton was president and consequently we had hundreds of visitors come and every year and they would come and they would say, this is great.

Bob: These kids are so engaged. This is awesome. And this would never work in my school, mostly because they looked at most, almost all the kids were white. And Marin County is fairly affluent and I just got tired of hearing that. And so I didn't believe it. So I helped co found a vision education, which you mentioned, and we created redesigned high schools, now middle schools, they use the same principles of project based learning and.

Portfolio assessment of the students work. And in fact, we got even better results. I mean, we have this last year. I'm still on the board there. This last year, we had 90 percent of our kids were accepted to a four year college. And this isn't Oakland, Hayward and San Francisco. They, uh, we have an 85 percent persistence rate through the first year college and a study by Stanford.

The students said one of the top reasons why they feel like they're successful in college is project based learning. Maybe talk to us a little 

Deep: bit about what doesn't work when you don't have project based learning. And I mean, I think, you know, any of us, those of us who have raised some kids. Can probably speak to it.

I'll throw something out there. You know, I think like one of the things that the curriculum based education system have struggles with is, you know, kids don't know what why they're learning what they're learning. And it seems like a huge problem and frankly I went all the way through in undergrad and even into the graduate curriculum in engineering, and it's not always obvious how to stitch up, you know, the classroom with.

Thank you. With the why and project based learning always just seemed super intuitive to me. And it seems like such a slam dunk as far as getting kids to care. And as far as both of my kids just love their projects and weren't particularly stoked about the other stuff. And I'm wondering what is it that's failing and what are some of the real.

Count like what's happening here? Why did we even get to this world where we have to talk about project based learning? Is it because, you know, curriculums are sort of so demanding and so like time consuming that there's no room left for project based learning and 

Bob: like what's what's going on there.

I'll give you a couple insights. I mean, I think that the traditional way of learning requires quite a bit of compliance and to just suspend Why does it matter? It's like, it's eat your spinach. It's good for you. Don't ask, just do. I don't think it ever really worked well. And I think we sorted a lot of kids out of educational through dropping out of high school, not going to college.

But we had an economy where there was a lot more jobs that were available that you could actually survive and maybe even have a family. Well, those days are pretty much gone. And I think the irrelevance. Now, you know, one of the things I always talk to educators about is like the world, like even if you go back like 10 years ago, the world has changed dramatically in the last 10 years.

But in the last hundred years, the world is like completely different. But if you walked into any classroom in any school in the United States, it would probably look very similar to the experience that a kid had in a school in the 1920s. Yeah, I mean, 

Deep: our whole educational system is like an 

Bob: industrial area.

Yeah, it's, it's, it's absurd. We were, I've had conversations with a AI company. It's called TeachFX. And they have an app that allows a teacher to turn it on when she starts her class. And it will analyze who's talking and how much. And they'll give her a recording or like a graph and she can see what happened when she asked the right open ended question where the kids were actually having dialogue and zero premise in their case is like.

Just having kids talk is a good thing. So we are talking to him and he goes, yeah, we did this study over the pandemic. They had a hundred thousand hours of recordings. They found out that on average, the amount of student talk that happened in classrooms across a hundred thousand hours. Six seconds. Six seconds.

And it was, I assume this 

Deep: does not include them just talking to each other behind? No, like in instructional, 

Bob: like an actual, like in an actual, well, but it would include dialogue process. Okay. We want you to go into a small group and talk about this topic, then you would have that. But kids, it's a, it's a teacher driven curriculum process where teachers talk direct.

And in fact, in some ways, I've talked to some folks in schools recently because of the pandemic and LMS is now you have kids, they go in and they all have laptops. So the teacher says, log in your assignments there. They get on the computer and they just start doing whatever the worksheet is. While the teacher walks around and make sure everybody is compliant, project based learning is, in a sense, the antidote to that thinking about it's inherently is about the why.

I mean, you frame it as, you know, in our framework, it has to be. sustained inquiry. It has to be authentic and relevant to the student's life or to the world outside of school. Uh, the work is public. It's not just for the teacher. You're thinking of a broader audience. You're asked to reflect on your learning.

Often you get you revise and critique based on feedback from outside the classroom or maybe inside the classroom. So it's much more active, not only in, you know, people would say hands on, but it's really minds on. The other thing I've been thinking a lot about is people try to think of project based learning is not being academically rigorous and what my when it's done well.

You actually, it's very intellectually challenging, but it doesn't begin there. It, it's like when I said like, eat your spinach, it's good for you. It doesn't start there. You end up eating the spinach because you want to, because you see the relevance of it. And, and ultimately it's actually more academically challenging than any Yeah, I mean, any worksheet you'd ever ask a kid to do the, the rigor argument is, 

Deep: it's odd, right?

'cause it's a function of how much you dig in. But. You know, every every PhD student on the planet is doing project based learning. I mean, it's it's by it's by definition what you're doing at that level. So it's not so much the construct. It's the how do you incorporate the prescribed curriculum aspects into the projects in a way that lets you check things off in the way that whoever is coming up with these curriculums like how when they Bye.

Bye. Want to basically say like, hey, you know, in order to eventually, you know, complete. A differential equations class you have to march through this, you know, whole sequence of stuff for 12 years or whatever 13 years and keeping track of all of that and then aligning that with project seems like it is a task like somebody has to figure that out and figure out how to do it.

And somebody has to set the rigor level, you know, and like how deep you're going to go. I mean, you know, even with the simplest simple kindergarten project, you know, if you gave that to a physicist, they could go pretty deep with it with most, you know, most of these topics. Right. So it really It's just a function of like, how do you do that?

So I'm curious, like what kinds of projects are we talking about? Are we talking about, Hey, we get a, you know, you've got three months to cover the court and the, you know, the quarter, the semester's worth of stuff. Um, we're going to give you two, one week projects to kind of like sprinkle in there in downtime.

Or is it more like, Hey, we're going to radically rearrange this thing. And we're going to do a hundred percent of project learning, but we're going to fill in, uh, you know, the curriculum to like, you know, to, to checkpoint it. Like how, how does it work? Maybe tell us a little about what PBL works does 

Bob: too.

So one of the things PBL works, we, what we say is use project based learning as the main course, not as dessert. So the way you described that would be like, the first example was more like, Oh, we're going to learn this. And then you get to do a couple of projects. In PBL works, we're known for setting the, basically the standard for what high quality project based learning looks like.

We call it gold standard. And there's two parts of it. There's the design of the projects. Um, and then the other part is the gold, what we call the gold standard teaching practices. We work to build the capacity of teachers currently to do both. We're moving in a direction of building more curriculum. So to your point of like being able to codify and go like, Oh, these, these projects are hitting these standards.

And here's an assessment tool that is very aligned to the standards that you. Set out to do, but you learned in a project way through inquiry and to your point, like when it's time to have kids deal with, you know, the core knowledge that they need or the skills. Then you're going to use some more, maybe what might look like more traditional teaching strategies, but they're in the context of this application that kids are working towards.

I think that's the biggest thing like we think of assessment is three things, knowledge and mastery of knowledge and skills. The current way of doing that, like giving a kid a test to know whether they have it is fine, but what we don't ask them to do is demonstrate their understanding of it or apply that understanding.

That's where PBL comes in, and we develop rubrics and other tools to assess whether they're applying it in the proper way. And then the third, which is almost never done, which is actually the most powerful part of project based learning and learning in general, is what we call reflect, but really is metacognition and asking kids to think about what did you learn and how would you transfer that learning into a new and different setting, and when did things not work right, and how did you adjust, and what will you do the next time, and what were the successes, and how will you make sure you repeat that.

Those are the, and you. Most educational experiences is one. It's a one dimensional. I tell you something, you tell me back, check the box, move on. But the real learning happens like we think of as a triangle, but down on the bottom, bottom sides. So we, we're moving to a place where we're going to build the curriculum, so we can focus all our efforts on helping teachers get the capacity to teach in a different way.

Deep: So do you like. Schools and districts, etc. Do they need to, it feels to me like they're kind of separate but related like you could come up with projects. where you maybe like take a break to go learn some some rudiments like for a while something out of a rubric in which case you could swap in anyone's curriculum like if it's I don't know you're doing a project on um I think on one on your website I saw one where they were You know, they're designing a house.

A bunch of students were designing 

Bob: a house project. Yeah. Yeah. And, and 

Deep: you could imagine where, where they sort of say, okay, we're going to design a house, but we need to learn some basic math. That's going to help us figure out square footage and, you know, maybe some electricity consumption, like whatever.

And then all of a sudden you, you go and you say, okay, well, that covers. Point A to point B in Singapore math and point, you know, C to point D in Saxon math. And, you know, like you could, you could imagine that kind of a thing, but then you're saying you have your own curriculum. Is it, am I thinking about it the right way?

If I'm thinking about it as you could swap in other, you know, curriculums. 

Bob: Usually what the curriculum is mapping backwards is from standards that are agreed upon. So the math standards, they would be the common, mostly common core standards that are. In most states or there's the next generation science standards that are agreed upon by like about 21 states and anybody who's creating curriculum is aligning to that.

And even teachers when they're creating their own are usually asked to say in this learning experience, which standards are the kids learning and how will you assess that for their knowledge? The tip of you know, what goes on 95 percent of the time is a teacher talks at a kid shows something. And then ask them to regurgitate it back.

And then they repeat that over again. There, I call it the myth of coverage. There's this belief that if you tell somebody something, they now know it. It's just not, I mean, and everybody intuitively knows that's not true, but teachers will tell you, well, I can't do projects because I have too much content to cover.

And I was like, well, you covered it, but they didn't learn it. The other there's research that shows that when kids take a test. Even the kids who do well on the test, if you give them that same test in six weeks, they'll fail it because they only learned it. It was, it was short term memory, not, not embedded.

So what happens in projects is you'd cover less the kids deeply own it. So when they go to take a test six weeks later, they still know it. Yeah, because it's something 

Deep: that meant something to them, right? Like, it's things they go home around the dinner table and talk about. Oh, yeah. And it's something they can get excited about, right?

It's not an abstraction in a math text or in a chemistry book or something. It's like an actual process. Need help with computer vision? Natural language processing, automated content creation, conversational understanding, time series forecasting, customer behavior analytics, reach out to us at Xyonix. com.

That's X Y O N I X dot com. Maybe we can help.

So then when you describe your curriculum and your projects. Are you thinking of it like, hey, I'm a school district, and I'm going to take all of the 3rd grade and go totally onto the system, which means everything's project based. Um, but on top of that, every project covers some part of part of what we would think of as a standard curriculum.

Bob: Yeah, we think about it a little differently. We, we think about it like in two ways about what's the minimum dosage, if you will, of project based learning that a kid needs to start developing a sense of agency. This application exercise get engaged. We look and then what's what's in the zone approximate development for teachers, because.

Uh, and we've found like two projects a year, like if you can get teachers to minimally do a project in the fall and a project in the spring, and then you think about that over the course of 13 years, that's 26 projects of increased complexity, you're, you're on the way. And it's enough that kids can start to think of themselves differently as a learner.

PBL is as good for the teachers as it is for the kids. I mean, the teachers start to like it and our hunch is that they'll start to do more projects. Getting a teacher to switch over to a whole PBL curriculum, it's a very difficult change. You can start schools. From scratch, possibly they're all project based learning and build capacity that way, but I haven't seen it really work to go full PBL.

It's it's a part of the challenge is like most teachers haven't ever experienced project based learning as a learner. They didn't they never did it as a student. They didn't have Mr. Cooper in the 5th grade. So, yeah, or maybe they did some 

Deep: reports and some, some stuff like that. Yeah. Or they did projects.

Bob: They did the state project, but we, we call that a dessert project. That's not a lot of people did things that were projects where they had to go home and do it. And so if your parents were around and they're engaged, then you got a great, you know, maybe you got a good grade because you're, they did it for you.

But, uh, science fairs come to mind, science fairs, like California, we have the California mission project. Um, that's all done at home. You can go down to the local art store and buy a mission and put it together. 

Deep: If we think about the higher grades, though, you know, like in high school, for example, the International Baccalaureate program, that's something that both of my kids went through.

They have, uh, some, some really, um, significant things, at least in the, I think in their diploma path where they have, you know, where the students have to really do extended projects. And they're, they're actually pretty in depth and they feel more like, you know, you know, like, uh, an actual thesis or something.

Bob: Yes. We, IB is, we call it like PBL adjacent. It's not quite gold standard project based learning, but if we had an IB program here in my community, I would have sent my kids to it as well. It's, it's far more deeper learning. And it's inquiry based and performance based project based learning gets more into authenticity and and public products that you're working for an audience, especially in the older grades beyond the classroom or like the tiny house project that you saw the even the fourth graders or third graders.

They're presenting their models to parents who came in to be consultants or other teachers or clients. It's it's amazing. I, the one thing I found out right away is like, as a teacher, when you have a public product and you move the audience beyond the classroom, kids work so much harder, they're, they don't really, kids are for whatever reason are okay.

Failing with their teachers and their parents. I always like to think it's because they feel loved, but then nobody really likes to look stupid to people they don't know. 

Deep: So this is an AI podcast. So let's maybe. shift gears slightly and talk about what's the AI angle here, because I can imagine a bunch, right?

Like I can imagine for one, um, you have a catalog of these projects. So in the process of designing projects, there could be a lot of, you know, AI leverage. I imagine in the process of taking a project. And like sort of transforming it to a particular grade and mapping it to specific sort of standards concepts.

There's a lot of machine learning and AI potential there. And then I imagine there's quite a bit of potential on the assessment side. So being able to assess. at the project level and maybe, you know, at, you know, at the grade or, you know, at this, at this, uh, across a set of standards or whatever, there's a lot, a lot, a lot of role.

And then there's kind of the whole tutoring or assistance angle, you know, to like help a student through a sequence of physics or math or, you know, something. Um, because if you're in this project based approach, inevitably, you're going to sort of spend some time covering the project stuff. So, you know, facilitating the independent time and making that more efficient might make sense.

Like, are there, like, how do you think about about that at the high level with the role of AI in this project based world? 

Bob: Yeah, we were thinking about it. I mean, we're probably spending most of our time thinking about it on the assessment side and I'll maybe go a little bit deeper into that in a sec. On the design side, what we've seen, we've been messing around with it, some folks on our team.

And it, and right now where it's at, it can generate some really interesting project outlines and ideas and concepts. Which if you're an experienced project based learning teacher, you could probably then map that could really help you map that out. And the AI could help you make sure you're mapping back to those standards we were talking about, um, because you could even start there, like design me a project that has inquiry that's hitting these common core standards.

If you're an inexperienced teacher to PBL, I don't know how helpful that is. But as maybe as we grow and we get more and more experienced teachers, that's going to be a really great resource for them. I think it can also right now, you have a project and you've outlined your project and you know, there's certain instructional activities that you want, like things that need to be learned.

AI is definitely going to be able to help you find the right lesson plans that tie to the actual knowledge and the way you want to do it. But I, I still think these are more advanced. teachers who are able to capture that right now. I think the big thing is, it'll be helpful for folks like us who are designing for teachers, but that's a small minority.

Like, so what's the structure like here? 

Deep: Are you, um, leveraging the community to design the projects themselves? And then you have another, probably a lot larger percentage of the community that's just consuming the projects and using them in class or, and your team is like kind of juicing that and facilitating the design of the projects.

Or does your team take on the burden of designing projects on their, on your own? Or like, 

Bob: how, how does that work? So our, our, our process right now is to do more, more of the, you working with teachers to help us generate the ideas and the concepts at a high level, and then take their good work and use our team's expertise to actually build out the full project unit and then bring it back to the teachers and say, what do you think?

Refine it and then say, now teach it and let us see what happens. And then refine it again, lots of iterations, but working with real teachers or with real kids. So, sorry. 

Deep: So you must have some, some teachers that are maybe just want to consume a pre designed 

Bob: project. We've done some research on this and like 30 percent of teachers.

We call them by the book. They just want their book and we want to give them a new book, a new curriculum that's better. And then you have about another 60 percent or a little less than that. Let's just say 50 percent that are modifiers or cobblers. They like to find other things and put them together or they'll take our project and they're going to adapt it.

And we, we actually tell them the places where we think it should be adapted for their community, the kids, the whatever's working for them, if they know their kids well, and only about 20 percent of teachers are actually want to design. It's, it's a lot of work. It takes a lot of time. It's quite fun. I love designing projects and curriculum, but, um.

Most teachers are not and they want they want some curriculum, but they want it to be curriculum that engages kids and bring some joy into the classroom. So that's where we come in. I mean, those are still 

Deep: Really high numbers as far as user generated content communities go like generally on the web.

You're looking at like, you know, 1 percent actually create if not less 3 percent amplify. So 3 percent are like the retweeters, if you will, and 97 percent are just the passive scrollers with their tongue hanging out like reading. So like I mean, you know, 50 percent cobbling is still pretty serious, uh, engagement and 20 percent are willing to participate in the design of the projects.

I mean, that's that speaks to the interest level and maybe the fact that there are actual. I don't know paid employees who are supposed to do this and feel like they should. I mean, 

Bob: yeah, I think I think it's also a factor of like a lot of teachers don't have aren't given a lot of resources to teach. You know, they have to figure out how to do it on their own or by the curriculum.

And then the flip side is. School districts will purchase and then mandate and actually monitor that teachers are actually following the script and the pacing guide. There's a mix of some places that are like so prescriptive that they truly take the joy out of teaching and learning. And then others that don't give teachers anything and they're They're, they're drowning what we're trying to do is figure out solutions to make it easier for teachers to do the type of teaching and learning and assessment that brings joy to them and the kids as well as learning.

So I'm, I'm 

Deep: like trying to think what your product looks like maybe today and maybe tomorrow. I don't know, but I mean, I'm envisioning like you're a teacher, you wind up in the system, you're, you're putting in some information about yourself, the grade that you're teaching, you know, there's probably like a simple search box to try to find projects that exist.

Um, there might be some default ones fed to you. Once you grab a project, you're deciding, do I want to cobble or make my own? If you want to cobble, uh, meaning like adapt that, then I'm envisioning a smart editor where you grab a project and maybe, you know, it's design a house, but you have a specific angle and, and the, and the, and AI tools, like facilitating the construction of that adaptation.

If you're trying to design one from scratch, then I'm envisioning like, you know, a lot of. Kind of orchestration to like what what it needed to like help them design a project that makes sense. So that would be like some large language model interaction to like actually generate the description of the project to actually come up with the concepts the ideas, have them kind of pick and choose and narrow down, probably some more standard searchy stuff with clever ranking to like figure out what curriculum stuff, you know, fits in there and then maybe some massaging to like get an anise.

To get it into the project. And then when it's all done, you know, they like, try the project, you track it. And now there's other teachers that are sort of grabbing onto that. So you have some popularity that pushes popular projects up and down and, uh, and then ultimately, though, if you know, if you're successful and people, you make this nice ecosystem of teachers creating great projects that, um, if you can start attaching like student performance metrics to the projects, Um, Then, you know, you start getting to an interesting place where maybe it's totally fine for teachers to just consume and because they're consuming the best design projects for that, you know, world they're in.

I don't know. 

Bob: Is that, is that sort of in the world you're in? That's about, you know, you capture where we think we'll be probably further in the future. Right now, we're starting with like, let's make sure teachers get high quality projects that they can learn to do PBL. And then. We'll start to see we'll start to build that community base, where all those things start to become possible and we can start to grow out more projects and figure out a process for curation and qualification, because, I mean, this is my point of view like I.

I don't want to have a low quality project that leads to like not great student learning, but it's because it's whatever reason it becomes popular because kids like it a lot. Yeah, so we're going to be a little bit more heavy handed on those things, even the discussion groups and opportunities for collaboration, making sure we always have a moderator.

And we're stimulating conversation and we're getting rid of the misconceptions, calling them out, um, in the place. I do think, you know, where, what I'm really excited about is the opportunity in the assessment piece. One of the biggest challenges for teachers is, especially in middle school and high schools, is the sheer number of kids and the amount of work.

I do this exercise with school and district leaders often, I'll just say like, When, when a teacher gives, let's even say a traditional sort of academic project, like a research paper and a sophomore year, and it's five pages and you have a clear standard and a rubric to assess it. I mean, how long is it going to take that teacher to assess 150 research paper?

I mean, it's going to take about 20 hours. Teachers are working from eight o'clock to four o'clock, and many of them have to go home and take care of their own families. They get, if they're lucky, they get a one hour prep period in that day. So when's that extra 20 to 30 hours going to come to do that? So teachers don't assign those types of tasks, or if they do, They just collect them and the kids really never get any good feedback on it.

Yeah. I mean, I think we've all had that, 

Deep: that teacher that, you know, gave you long papers to write in English later or something. And then you come back with a grade and nothing else. Which is completely, I mean, I, there's definitely value in just the fact that you wrote the paper, but it's quite deflating to get 

Bob: nothing back, right?

Whether you're a good writer or not. And then you're going, you know, yeah, I mean, I guess 

Deep: I've, I've, I've had, you know, like my, my daughter's heading off to college, you know, in a few days, but. You know, like I remember when she was in maybe ninth grade and she's getting all these papers back from her English teacher and they have like no comments on them, and they come back two and a half months late.

And I'm thinking this is, this is absurd, and she's getting some high marks on them I'm like, look, this is your misusing semicolons and there's like a million things that's going wrong here and nobody's telling you about it. And these are things that are really straightforward to get GPT 4, you know, or a 

Bob: quick lens to get, I watched it today, deep in 30 seconds, the open AI tool, uh, assessed as piece of student work on a rubric and gave the student actionable feedback.

And a, and a score and it, and that, that student could then revise that based on it in real time, like right now, not wait, have to wait like two weeks, like as a school system leader and school leader, you want your daughter's class to have the same standard for assessing that essay is the next class.

Well, the teachers need to get calibrated. And it's very hard to do, but the computer doesn't.

Deep: We've, we've been designing a lot of these systems where we, um, you know, for clients of ours, where we, we have some kind of rubric, you know, like some kind of editorial construct that has to get applied to a body of generated something, text or conversation or something. 

Bob: And, and these models are. They're quite good at it.


Deep: they can, they pick it up. They're much better than most, the vast majority of teachers that I ever interacted with. So I think you can get it 

Bob: to like a better 

Deep: place. Like I talked to folks a lot, like, um, so one of the things we do is we, uh, you know, we have writers that write stuff for us. And I know a lot of people are, you know, we don't want any marketing writers that use chat GPT, which seems like the exact opposite thing that you want to do.

So my, my thing is like, I don't want to ever have a writer. That doesn't start off and have a robust way of thinking about how they're going to use GPT 4. I don't want 3. 4. And I want them to explain to me their methodology. And then like for our editor, the feedback that you're giving is not domain specific enough, and it's not digging in deep enough.

This is what I want you to ask ChatGPT. It's like simple stuff, but the feedback ends up being spectacular. Why is this article boring? Question mark. Boom. Great feedback comes back like, and everybody's article is going to be boring at some level, you know? And then another one is like, rewrite this passage using way more examples and, uh, real life companies with real life quotations.

And it. Fakes the quotes, but it tells you it faked the quotes. But if you take that text and search on it, you immediately get real quotes from the people, but it picks the right people to quote from and the rewritten versions are like 100 times better than the people. So I feel like educators are going totally down the wrong path of banning, 

Bob: you 

Deep: know, LLM based assistance.

I feel like it's 100 percent the wrong direction to go, but they don't know what to do, and they don't usually have the time to figure out how to, like, First of all, they'll fail miserably. Even if you ban it. I mean, every college that's banned it has every one of their students using it anyway, and it's easy to beat the spoofing and I can prove to you that you can't ever build good plagiarism detection.

I don't know. I think it's just like a new beast and everyone's scared of it. But in reality, we should be totally embracing this stuff because I think you can get kids to 

Bob: perform so much better. Oh, yeah, they could. They're now more engaged in what they're doing. And, and using it as a tool. I mean, that's, I've been from the very beginning, like the last 30 years, like to work with educators.

It's like, this isn't about learning technology. It's about using the technology as a tool to advance the learning that you want to do. Well, this is just another very powerful one. You know, I do think that there's a, one of the other applications is, You know, AI generated tutoring and coaching. I'm still like thinking about that, pondering it.

I think that in a project based framework where a student is motivated and is like, Oh, I want to like, Learn about this and get better at it because I'm going to apply it in the, I'm building a bridge for my physics class and I want it to be the strongest. There's going to be a test. Engineers are coming in and going to do the test, but I need to understand these concepts better.

I'm going to go. I have access to this learning opportunity that would probably a kid would get into it, but where I see a lot of interest in philanthropy and funding AI generated tutoring is for kids who are already disengaged and struggling with. math or reading. It doesn't give the kid the why. It's not a joyful experience.

It's not connecting them into what, you know, why learning can be a very engaging generative experience. Instead, it's like, oh, you lost again. You get to go sit on the computer when everybody else gets to go and do something fun. I mean, I feel like A 

Deep: huge chunk of the momentum in the institutional educational arena for the last 20 years or so has gone into this kind of standards rubric formal assessment thing.

And I feel like what you're honing in on here is what I've always thought was missing, which is the motive, like the best teachers in the world that I've, I've seen, and the example I use is like so I recently started learning guitar. And I have this guitar teacher. So before I started taking guitar lessons, uh, I would sit down, I'd start YouTubing some stuff, I'd come up with something, I'd develop all the wrong hand techniques and tactics, and then I'd dink around, and I'd produce mostly garbage.

And a couple years of that, I pretty much got nowhere. So then I, I decided to take lessons. I get this guitar teacher. And the thing that he did in the first hour lesson was he got super excited about something so mundane and boring that like just simply just getting one chord to sound clean once that what it did was it it kind of it just motivated right so now like It's like, Oh, I got somebody excited.

Now I'm like in social debt to this person because I spent an hour with me. So next week I want to go do more. Like there's all these other things that happen when you address the why. And project learning, like, like I remember, you know, I was in engineering school and I, I was very kind of privileged. I, early on, like my freshman year, I.

Sort of hooked up with a research professor who let me basically just hang out with a bunch of grad students in a research lab for four years and do my own research and publish papers and do stuff. So I learned kind of backwards but project based so I didn't really care at all about my coursework. I only cared about my projects, but it turns out that I was limited in the projects I could do what so I would sit down with, you know, this advisor of mine every semester and be like, Hey, I'm struggling with this, like, what class am I going to like learn that's going to help me be able to read these papers and figure this out.

Like, oh, as soon as you get your signals and systems class, which, you know, I think right now is scheduled for next year. But if you take this thing now, you can get to that within like three months, I do. And all of a sudden, like this whole world would open up for me. Yeah. So I was always like, I always loved 

Bob: school and all my 

Deep: peers were just kind of like, whatever, let's just go get a job and get out of this crap.

And it was just, You know, and I think it was just because I had like this backwards approach that I think shouldn't 

Bob: be backwards. No. It's a way that that's the way we learn. I mean, it's funny, like when I'll ask educators in a, or sometimes I'll do it in a keynote. I'll say like, recall a transformational learning experience that you had in high school.

I'm working with high school teachers. I tell you it people can't remember one. I mean, they block out high school. Maybe they'll tell you about something outside of school that happened, like a coach, student, music, drama, uh, journalism, but it's a big black hole. And people actually, I think. Are trying to forget.

Um, and it doesn't have to be that way. It can be. I mean, it's such a transformational time as a human being that it's unfortunate that most of the most transformational learning we spend during the school day. Yeah, you're in there like 80 percent of the day, but you do the learning that happens. And the little slices outside, like maybe you went to camp or the different, the different things we just discussed.

Well, I was talking about, it's like, you can actually be intentional to create project learning experiences that recreate all those same things, all the same emotions and feelings that then you, you know, students that have gone to my schools, you can ask any of them. They'll tell you about three or four different.

You know, transformational learning experiences. They had in high school. So we're almost out of time 

Deep: here. Thanks a ton for coming on. I want to end with like 1 question that that I like to ask fast forward for me 5 or 10 years out. And the kind of core question I have is. Do we have a shot in the hell, at least in the US or in the Western world, maybe more broadly of radically inverting this educational paradigm around project based learning, which I would personally love to see.

But like, I'm sure some people wouldn't. But like, that's like, is there a real shot at that? How would that happen? And what would be the role of AI in making that feasible and advantageous and, you know, or not 

Bob: and ruining it? Well, last week, my board approved a new strategic plan to build out a digital solution to provide our curriculum with embedded professional learning opportunities for teachers and using generative AI assessment tools to assess those projects.

Right now we work with about 10, 000 educators a year doing basically in person or synchronous learning. Our goal over the next five years is we believe by building a digital platform to do that. We can we're not giving that up. We it's good work. We can probably double it, but we can get to there's about 3 million teachers in the country in the next five years.

We think we can get to about 300, 000 of them. Um, and now we're at 10%. So we're still scratching the surface. But I think it's, uh It'll, it'll be a flywheel type of experience. Like we can get it going. So I'm very hopeful and very optimistic that the tools that are a lot of the things that I've been wanting to do in the last 30 years to try to make project based learning ubiquitous.

There's actually now we're living in a time where there's the digital tools are maturing enough that I think we can make some good headway on this. And do you 

Deep: think like outside of outside of your company, you know, when you when you talk to government agencies or folks who, you know, really hold the Curriculum strings like is there an ear for this approach for like changing things very, very significantly?

I feel like in higher ed like capstones have had a huge impact, you know, like almost every engineering and other departments, you know, seem to be like implementing capstone projects. I'm curious if you're seeing like others recognize this need to 

Bob: Yeah, our work is warm. Yeah, it's it's in K through 12. I actually see a lot more transformation possibilities and happening in K through 12 and higher education, but You know, even if we just start with the national policy and the Department of Education, I mean, they explicitly call out project based learning as a strategy and as a need in our country, I think, especially coming out of the pandemic and the lack of engagement and That you've seen, and especially in secondary students.

I mean, they're missing. They're just voted with their feet. Um, people are, we 

Deep: lost like 8000 students in Seattle. 

Bob: They just people are looking are trying to figure out Simultaneously, how you build the learning loss, but also how you engage kids. There's an openness that also that I haven't seen, um, and an interest all across the country.

I mean, we're, we're, we're working in the state of Kentucky and our goal in partnership with the Department of Education there is 450 schools, a third of the schools in, in three years. So we're about two thirds of the way there. So I'm, I'm more optimistic than ever. I feel like, There's many years like from 2001 to about 2010 where it felt like the wind was blowing hard in our face and it was going to be, it was tough going, but it's, I don't know if it's in our back, but it's not, it's, it's a slack tide right now.

Well, that's, 

Deep: that's better than one going against us. That's good. Well, thanks a ton for coming on. I think, uh, it was just awesome talking about this, and I hope, uh, hope we get a lot more students learning from a project vantage and engaging hire. That's all for this episode. I'm Deep Dhillon, your host, saying check back soon for your next AI injection.

In the meantime, if you need help injecting AI into your business, reach out to us at xyonix. com. That's x y o n i x dot com. Whether it's text, audio, video, or other business data. We help all kinds of organizations like yours automatically find and operationalize transformative insights.