Week in Edtech, 9/22/23, Part 2: Deep Dives with Toddle Learning and Project Read

[00:00:00] Alex Sarlin: Welcome to Season seven of Ed Tech Insiders, the show where we cover the education technology industry in depth every week, and speak to thought leaders, founders, investors, and operators in the ed tech field. I'm Alex Salin. 

[00:00:20] Ben Kornell: And I'm Ben Cornell. And we're both EdTech leaders with experience ranging from startups all the way to big tech.

We're passionate about connecting you with what's happening in EdTech around the globe. 

[00:00:32] Alex Sarlin: Thanks for listening. And if you like the podcast, please subscribe and leave us a review. 

[00:00:37] Ben Kornell: For our newsletter, events and resources, go to EdTech insiders.org. Here's the show.

Hello everybody, and welcome back to EdTech Insiders. It's another great week. I'm your host Ben Cornell, alongside Alex Starlin. We are excited to bring you the headlines and news, everything going on in the world of EdTech. But first, let's cover everything else going on at the pod newsletter events, man, September and October is gonna be huge.

Alex, tell our audience about it. 

[00:01:12] Alex Sarlin: So we've been sort of going really crazy on the pod in, in a really exciting way, talking to a ton of amazing guests. This week alone, we have an episode with John FAA of Pearl. This is a really fascinating tutoring platform that's going B2B with all these tutoring companies.

Really interesting interview. We just put out our last week at EdTech and we have Bodo Hernan from knowledge. Fascinating guy. I mean really one of these just out of the box thinkers. Self-trained, learned ai, learned how to make cybernetic limbs, and is doing this really, really moonshot kind of stuff with knowledge.

And then we just have an episode just this week with James Grant who runs a really big British tutoring company that is coming across the pond to the US and has a really cool, very relationship driven model. So we we're just getting episode after episode. Amazing person after, amazing person. Check that out.

What's going on with events this month? 

[00:02:09] Ben Kornell: Well, there's ample opportunities for our EdTech community to connect. We are going to be in full force at New York EdTech week coming up, and on October 3rd we're co-hosting with Magic EdTech, a rooftop happy hour. It's at versa. Please reach out if you want to come.

It's gonna be a blast, beautiful rooftop location right in the financial area of Manhattan. Beautiful views and just great conversation with all, um, EdTech, founders, CEOs, technical leads. It's going to be great. And then on October 11th we have our AI founders forum, which. At four o'clock, there will be a wide open Bay Area, happy hour, anyone can come.

It's in the old Zynga building. We have the entire first floor. It's like a private bar. It's gonna be awesome. Co-sponsored by Common Sense Media. Very, very excited about that. And then of course, we have our online conference, October 26th. It's really an all day affair, and for those of you who can't make it to one of our in-person events, this is an awesome opportunity, not only to hear from great speakers and panelists, but also to have time for dialogue, discussion, and networking.

Each panel essentially has 30 to 40 minutes of panelists time, and then 20 minutes of networking, connection, questions, dialogue. It's going to be great just like every EdTech Insiders event. So we've got it packed for October, Alex. Yeah. 

[00:03:40] Alex Sarlin: Back to school season. We're leaning in. So let's talk about what's going on in the world of EdTech.

Let's start with ai. I know we do this a lot, but there is really cool stuff happening in ai. Do you wanna kick us off? 

[00:03:52] Ben Kornell: Yeah. Well, for those of you who read the newsletter, we had a new episode or new edition out, which really covers the implementation layer and one of the. Key findings that we shared was really a report from Reach Capital around trends from 280 plus generative AI EdTech tools, and they really focused on kind of how do you segment the market, what are the different categories, and then who are the leading companies.

It was felt like a milestone report it did in that. Now we're starting to see some definition and some maturity. Where I found a lot of excitement was really around the teacher enablement tools. There's a bunch on personalized learning and tutoring, but it did seem like the kind of. Speed of the space was matched by the funding, over $350 million just in personalized learning and career and skill development.

So tons of generative AI investing going on in this space. And almost, you kind of have this like generative AI ed tech investing bucket. And then you have your non-AI ed tech investing bucket, and it feels like the generative AI one is going to surpass the non generative AI bucket. So those were trends that I was looking at.

Um, several other big articles. But let's start first with the seminal reach report. What were your takeaways? 

[00:05:15] Alex Sarlin: Yeah, I mean, I thought it was really validating in a way they've looked at almost, you know, 300 different tools and because Reach has this AI Catalyst program, they're getting pitches for early stage ed tech companies that do AI all the time.

So I think Reach has a pretty good front row seat to what's happening in AI and education and some of the categories that they sort of landed on. I think were really fitting, it makes sense for the field and it makes sense for sort of what we've been seeing as well. There's this personalized learning and support that includes virtual tutors, mental health study tools directly for students.

It also includes. Copilots and teaching assistant tools, which we've seen a ton of. Then there's clear and skill development, content creation. We're gonna talk about all these things at the, at the online event at the end of October. But educational content creation, the ability to create text, images, videos, games, I.

Then use them as, you know, textbook material, as teaching material as interactives should not be overlooked as a core part of this space. You know, they have AI powered research, they have the plagiarism and integrity pieces, and then operational efficiency, which is what we've heard a lot of the EdTech companies focusing on.

As the picture becomes clear, they say, well, at least we're doing all sorts of stuff internally, and as we get together our strategies externally. It made a lot of sense. It was really interesting to see how many tools were in each AI powered study tools was the top in terms of number of companies. There were over 50 tools that were AI powered study tools for students.

Makes it a pretty crowded space very quickly, but I think we're gonna see it, you know, thin out as a few really take off. Yeah, I mean, I think they did a great job of really making sense of a very complicated, very fast moving space. Even just looking at, they also divide things by emerging funded and incumbents for each of the categories, which is fascinating because you see the incumbents, places like Quizlet, grade scope, Khan.

They've been around a while. Duolingo, Grammarly, right. Has been using this while Kahoot, teachable. These are companies we, we talk about a lot, but it goes all the way down to these emerging companies. Like we see knowledge on the map here, who we just talked to on the pod or Prof Gym, who makes these amazing educational videos with avatar speakers.

This spectrum is so interesting. You see what's happening, you know, the big incumbents that are moving into this and then this flood of new companies trying to break in and all these different angles. It's such an exciting space. 

[00:07:40] Ben Kornell: Yeah. A few things that really did surprise me were around the virtual tutor space was actually much smaller than I would've thought.

Seems like more people are going with the AI powered study tools, so that would be more like flashcards and study resources, which I guess I can understand because generative AI seems well suited for that. But on the virtual tutor space where we thought we would all have like an AI avatar teaching us things, that seems to be the lowest area.

I also was looking at language learning with 36 language learning apps, and EdTech has really struggled to show real market traction and IPO caliber companies, and yet language learning is the one area where it's just the monetization has always been great. It makes me wonder like, have we reached a saturation point on language learning or is there still a lot of room here with 36 companies going after it?

You have to wonder like how are they all, or even a subset of them going to find traction given how many big dog incumbents there are. So those were some categories that really caught my eye. And then the academic integrity space, which is really like catching people cheating. I. That was showing 9.5 million in funding.

If you had asked folks in November or December last year, where is the puck heading in terms of, you know, funding for this? It was around like catching people cheating and the compliance element of generative ai. And it seems to me like that's really fallen off the rails in part because the technology has shown limited efficacy and so the ability to kind of trust in ai like.

Plagiarism detector is going down, but it also seems to be hard to differentiate yourself because it's more of a commodity service. And so I'll be watching that space quite a bit because academic integrity, by the way, has been a huge space. I. For many years, and you look at some of the monitoring services and softwares, they've also done really well, but it seems like that may not be the kind of killer generative ai, you know, EdTech use case.

[00:09:59] Alex Sarlin: Yeah. You know, it's super interesting because in that space, you obviously have places like Turn it in that have been, you know, dominating the sort of plagiarism space for quite a while. You have copy leaks, which is rising in that space. GPT Zero is that company created by the Princeton student. But I agree, I, I'm actually very excited by the fact that this is not where everything's going.

I know you are too, but like the fact that everybody was talking about integrity and cheating, even students, that was their sort of main takeaway of what this stuff was. Oh, it's a cheating tool. It's a tool to use to write reports for you. That felt like the silliest take on this technology, and I'm really glad that this is not where all the money and all the funds are going.

That said, you know, we talked to Honor Lock, which does online AI and human-based proctoring on the pod, and one of the things that was really interesting, what they said is that you have the companies that are already deep in the monitoring space, right? Because all schools have. Firewalls and white lists and things that already monitor internet use.

And a lot of those companies have very quickly added on like Lightspeed, you know, AI detectors. Are students going to chat GBT? Are they going to any of these tools? Are they plugging in browser extensions? Various things like that. And I think that's actually part of why this space isn't blowing up, is that you already have some really big incumbents, not necessarily in the AI space, but in the monitoring, in the student integrity.

Let's chase students around as they go around the internet space. And I think they're trying to, you know, shift in, they already have, you know, they're in schools, they already have contracts, they're getting asked to do this. But I also think it's a really positive shift, which is that people are not thinking of Jet GBT and all of these amazing tools, these gen AI tools as just cheating tools anymore.

They're thinking of. Oh my God. They really could be tutors. Like Khan has been promising. They really could make amazing content. They really could be used to do incredible research like Elicit and uh, site are incredible tools. I think people's minds have really been opened over the summer and I'm like so pumped to see all of these different use cases being highlighted.

[00:12:01] Ben Kornell: So true. And let's also talk about what's changing in the space with teachers going in on generative ai. We highlighted a Walton Family Foundation survey in our newsletter. We also saw some stuff coming out from Magic School ai. What's the kind of changing. Mindset in the space that you're seeing. 

[00:12:23] Alex Sarlin: There was a, a really neat article in Wired Magazine this month about teachers going in on Gen ai and basically I think, you know, we're starting to see a few teacher facing tools that are starting to really get traction.

We've talked about Sizzle Magic School is, seems to be getting a lot of press right now. We've talked to Brisk teaching, but a few of these are just starting to be go-tos and you know, this makes sense. The school year has started, teachers are starting to look for this kind of, you know, what tools really work.

And I think there's starting to be a real embracing. The Walton Family Foundation found that 70% of black and Latino teachers are using AI technology weekly. That is. Amazing. And you know, we talk a lot about equity in AI and how we are gonna get this technology in everybody's hands. That is about the most encouraging statistic I can imagine when it comes to that.

I mean, again, we're beginning of the school year. It's the very first full school year after this tech appeared. The other thing we're seeing is that the A FT, right, the American Federation of Teachers, the Union is now developing best practices for teachers using ai. So, you know, we saw UNESCO doing this.

We've seen all of these organizations starting to lean in and say, what are we gonna do with ai? We also saw a survey saying about half of states will get to this later, but half of states have policies right now. Other half sort of don't. I really like that there seems to be a growing consensus that this is absolutely gonna be here to stay.

It's going to affect education. Teachers believe it. You know, principals and admins believe it and now it's just a matter of making sense of it and making sure it works properly. That is so big and I think we're gonna see the numbers rise very rapidly as the school year starts. Because once you have one teacher using these incredible suites of tools that make, you know, lesson plans, they make personalized studied materials, they make flashcards, they make quizzes, they make all sorts of things and they're all beginning to make almost all of it.

It's hard then for the teacher next door to say, you know what, I'm gonna continue to do all of this by hand. We talked to Todd this week. Remember how they said it was thought? It was so cool that one of the things they're building is the comments that go in quarterly reports for students. They're doing that with AI and that's something that takes teachers hours and hours and hours.

So it's saves teachers so much time, it's gonna spread like wildfire. 

[00:14:38] Ben Kornell: Yeah, I, I think that we've kind of crossed the path of like, is this gonna happen or is this not going to happen? I think there was this binary, like good, bad analysis and some people were landing on good and some people were landing on bad.

And in education it tended to be around academic integrity, around safety and so on. In the broader landscape it was, you know, are we gonna have ai, robot overlords? Is this gonna end humanity? And whether any of those things happen, I think there's a more of an understanding that. This is happening. It's basically inevitable.

The genie's outta the bottle, Pandora's box is open and there could be a lot of good, and now what we're actually seeing is the nuance, which is like, okay, so if we assume that this is where the puck is headed. How are we going to enable it for education equity? How are we gonna enable it to close the gaps we've been trying to close?

And then also unlock teacher capacity, which is probably our biggest constraint. Dan Meyer put out a really, really interesting article that also talks about how challenging that will be to create a copilot for educators. And you know, this idea of teachers doing. Different kinds of work in different places, in different modes.

One thing that I always come back to is this idea of modes of instruction. There's like direct instruction, which is whole group. There's small group, there's one-on-one, there's async. There's a way in which like actually building a full suite copilot tool is quite, quite challenging. Whereas building a copilot tool for a software developer is way easier.

And so as Randy Weingarten says, like, look, we're gonna put the, you know, guidelines out. That's a monumental thing for labor unions to be leaning into it. She also said. Generative AI is the next big thing in our classrooms, but developers need a set of checks and balances so it doesn't become our next big problem.

So I think you're starting now to have a much better, I think, really productive nuanced discussion about how can AI help educators? In what context can it help educators? And same for students and on the university level. I also think there's a way in which, because the student is really an agent of their own learning, and we've seen many, many new higher ed models, there's also a lot of thinking around, okay, how do we shift the balance of agency to the learner?

How do we create even systems and structures and stackability that respond to the learner? So it's, you know, we're in the great 23, 24 I I mentioned is the year of the gray area. It's also the year where it's just gonna be a big sorting out of how this is gonna happen. 

[00:17:28] Alex Sarlin: Yeah, and you know, I don't think it's a coincidence that there's more movement in the teacher space than in the virtual tutor space because a big part of it is what can you put in front of students at different ages?

Because gen AI is still a black box for many, many people. And I, and so, I mean, we're all figuring this out and it's happening faster than I thought. I wouldn't have thought there'd be almost any student facing tools this early in the year, and we're already seeing them, but the fact that there's so many competing teacher tools is, makes a lot of sense to me.

One of the things I loved about Dan Meyer's article and Dan Meyer for those. Who might not know, you know is, is sort of this legendary math teacher did a really famous Ted talk about teaching math in really innovative ways, and then has been a very close advisor, maybe even sort of in the team of Desmos, which is bought by Amplify.

They've done really, really innovative work in math and one of the things he said that I think is really interesting, he has these three things that he says made might keep a copilot, and he uses the GitHub copilot, like the coding tool as a metaphor. He's like the GitHub copilot has taken over the world.

It's like everybody's using it already, but. Is that gonna be true for these kind of teacher tools? And he's like, well, teachers actually do many more different types of work than coders, believe it or not. And there's actually a breakdown of all the different things they do and they're not in the classroom.

There's many, many different things, and that's why these tools are trying to spread and do many things. But he's like, there's many things. They also do things in many different places. They use lots of different tools and they need more than text. It's not just about text for teachers. They need all sorts of types of media to be able to help their kids succeed.

And I think those are good points. That said, I am still quite bullish on these teacher tools as you can sort of hear here, because I think that all of these things are actually conquerable by the tech. I mean, you're already seeing the teacher tools going from things like we do one thing, we make flashcards to.

We're a suite. We do everything you need. We do 10 different things. They're all interconnected like Tottle, or they're all built into the browser, like Brisk. I think these companies are moving pretty quickly, but it's still a great point. And what he says I, which I think is really interesting, he says, look, quote, if I were building one of these teacher co-pilot tools, I built a plugin that integrated with Google Apps for education.

A tool which has something like 60% of the K 12 market. I'd focused the co-pilot tightly on the needs of liberal arts teachers since they make the most use of text. And the co-piloting would happen in line with the teacher's existing work without any need to visit an external site and engage in round trip copy paste.

Very, very good product advice from a wise math teacher. He's, he's suggesting this going to liberal arts, but he's coming from math, so, you know, take note, ed tech founders out there, all you people building these suites. Very, very good advice coming from Dan. 

[00:20:12] Ben Kornell: Very specific too, one thing I love about the advice is, one, it's really important to understand that the user experience is not just what your website is.

That is not ux. UX is really the workflow that you experience and how you are creating a. Adoption loops in that workflow that deliver rapid time to value, like positive time to value to the user. And there is a way in which like an integration layer on top of everything that you do is the most seamless.

It creates the least like friction. And I've seen so many great companies that have had a great product and they're like, it's free. Why isn't everyone using it? And it's because it's not free. It costs a lot to rip everything else out and then use the free thing. So it's actually quite, quite expensive.

Now the other thing I would say is. There are some people who read Dan Meyer's article and say, I believe the absolute opposite of this. And I think there's also a contrarian opportunity here to go the total other way and say everything needs to be replaced. It is math, it is science, or something like that.

But I just think what is the danger of generative AI is that you can say yes and yes and yes and over and over and lose focus. And so everyone should read articles like this and, and like Dan Meyer's article, which we'll link, but take a strong point of view and either prove it right or, or totally prove it wrong.

I think there's room for a lot of that. I think there's room for both, but only when we have people taking an opinionated stance will the products really cut through. Whereas if it's just a peanut butter layer that kind of molds to every individual's interests, then you're gonna lose to the incumbents because the incumbents have so many other advantages and they're going to be able to peanut butter the AI better than you can.

Oh man. We could do the whole episode on that. And by the way, this is the kind of stuff on October 26th that we will be talking about in the conference. Let's dive in a little bit though. You know, we've talked about the what ai. Let's talk about the why and how and dive into K 12. So many important and incredible articles coming out this week and New Data First Curriculum Associates has a learning loss report with back to school data.

For those who don't know, curriculum Associates. Has this assessment called i-Ready, which is really just, it's everywhere in, in K 12, and it's a way of assessing students around their academic performance and reading and math and some shocking results. First, grades one and two students are performing below pre pandemic averages.

Okay. You know, I've heard this many times, but wait a second. Kindergartner, the grade one and two, they were preschoolers during the pandemic. So why are they performing below pre pandemic averages? So there's clearly an effect of COVID, even if you weren't in the school that was disrupted. Is it the teaching?

Is it the kind of socialization? We don't know. If you look at grades five through eight, 50% of students are reading on grade level, which is actually comparable with pre pandemic levels, but there's about a four to 10 percentage point reduction in math. So there seems to be in grades five through eight, those are the older kids they were probably in, you know, first, second, third grade during COVID, they're behind in math.

And then the inequities for black and Latino students continue and schools serving a majority, white students especially have bigger gaps with black and Latino students. So really, really important. Data there. There's also common sense, there's a clever report. What's some data that really caught your eyes, 

[00:24:14] Alex Sarlin: Alex?

I love seeing all of these really cool surveys come out with different aspects of the education and EdTech world and come out with data that's really interesting. Learning loss data. Very interesting. The common sense data is interesting. The thing that they did is very specific. They sort of dove deep into all of these different ed tech vendors and how they explain their privacy policies, how they say what is happening with the student data, whether it's being collected, whether it's being sold, how is it, you know, what is really going on there.

And they basically rank all of these different tools, whether they're clear at explaining or worse at explaining, and how that's changed over time. And you know, their main conclusion as far as you know, my read of their main conclusion is basically they're saying that almost three quarters, they're saying 73% of the industry should be disclosing in their privacy policies that they.

Are selling consumer data and are not disclosing that right now. So that's pretty large number. And it sounds like, you know, that's the kind of thing that common sense is always kind of a watchdog slash, you know, keeping an eye on the best practices and who's doing them in the industry. And that's a large number of companies to call out by saying, Hey, you're not telling schools that you're or for B2C, you're not telling, you know, parents and families that you're not only collecting some consumers data, but you're actually selling it.

I hope that's a little bit of a wake up call. I hope that Ed tech companies, you know, hear that and find this report and say, yeah, if we're doing this, we really should be clear about it 'cause it's not great. The other thing that they're finding is that almost half the industry is not in compliance with a, I think it's a California protection requirement that businesses disclose whether a.

Consumer's personal information is either sold or shared. So they're basically saying that almost 73% of the industry is not really disclosing this. Right. And about half is probably actually on the wrong side of the regulation when it comes to this and maybe, you know, is at risk there. You know, those are pretty high numbers and I think it doesn't reflect great very well on the ed tech field that we're on the wrong side of this, but hopefully common sense is, you know, raising the flag and.

Maybe privacy policies will be reimagined, especially in this, you know, AI age where people are starting to focus on them more. So that stood out to me. There were two more studies that I think we should talk quickly about. Ben, tell us about this clever report. 'cause that's really about EdTech and ai.

[00:26:37] Ben Kornell: Yeah, so clever. Very, very widespread single sign-on system that really allows for data privacy and protection, as well as kind of the connective tool and infrastructure for EdTech products, including rostering, et cetera. And so they have very large penetration and a very large survey catchment. Things that really caught my eye, number one is AI in the classroom, 89%.

Of teachers believe chat, GPT will have a moderate or significant impact on teaching. That's an increase from what we just reported last week in the Walton survey. The 89% of districts do not currently offer AI focused professional development. Also very aligned with what we talked about last week. What I also felt like emphasized some of the themes that we saw in the REACH report was teacher workload seems like the biggest opportunity for AI in terms of reduction of their workload, allowing them to focus on the teaching and learning, and kind of take care of the repetitive tasks.

And then there's a bunch of questions around diverse learners and supporting through personalization or through differentiation or you know, kind of AI assisted IEPs. How might we support diverse learners? The basic takeaway is the kind of fall has seen. A huge push of EdTech tools and AI infused EdTech tools with almost no professional support around how to implement pedagogically.

And then also, um, the State EdTech Trends survey from Set da. This is basically a group of technical leads across school districts and across states. They also had a 2023 EdTech Trends report. What were the main takeaways that you saw from there? 

[00:28:35] Alex Sarlin: One last comment about the clever report, just one thing, it was also done with whiteboard advisors.

I think that's just important because they are such embedded in so many different ed tech tools. I totally agree with you about the ai. One thing that stood out to me too, just before we move on, is the ed tech enthusiasm basically said that 79% of teachers say that that students are either as or even more positive towards EdTech than they have been, you know, pre pandemic or in the last, in the last year.

That is really exciting for the field because I think we're, what we're starting to see is. Post pandemic. You know, the investment dried up real fast, but it's becoming clearer and clearer. That tech really entered the classroom in a big way and people are actually pretty into it. Teachers and students are starting to be like, yeah, I like this, and it's starting to make sense.

And I think that dovetails really nicely with one of the things in the set that report, they asked about many things, but one was about collecting data on the usage and or the effectiveness of your ed tech tools. So basically saying, Hey, hey, state education technology directors, what data do you collect?

Do you collect usage data? Do you collect effectiveness data? And the fact that this is being asked in this way is great. Shout out to Carl Recana from Learn Platform among many others for putting effectiveness front and center. The cool news there is that within the last year we saw a shift of 10% going from 20% to 30% of people saying.

My state does collect both usage and effectiveness data. So that's a pretty big jump and we see a decrease in the opposite the the number of states that said, Hey, we don't collect any data on this, which is sad but true. That went from almost 50%, 47% down to under 40%. Now, the fact that almost 50% of people last year weren't collecting data on usage or effectiveness should make all our minds blown a little bit, but it's moving in the right direction pretty quickly.

And that's really, really exciting because not only are people sort of into the ed tech that they're using, but they're starting to actually get smarter about, you know, what are we actually using? What's the effectiveness? What's really working? What should we stop our contracts with? What should we double down our contracts with?

That is great news for the field. 

[00:30:49] Ben Kornell: Absolutely. I feel like we've done a lot on K 12 here, but just for our listeners to know, it really is astonishing that the space has changed at this rate with this kind of speed of adoption, but also optimism. And I will say that some of the data from the Set DA report, and by the way that was put together by John Bailey, who's also a friend of the pod, and just a really, really established and credible leader, it still shows that teacher shortage is number one issue for schools and districts academic outcomes.

Number two, equity number three, learning loss number four, and mental health number five. And so. Ultimately, where are people spending their time? At the district level, they're thinking about cybersecurity at the district level. They're looking for guidance related to ai, but don't really have clarity on how to implement.

And then ultimately, the kind of funding cliff that we've been talking about with kind of federal stimulus funds going away is really forcing them to recalibrate their priorities while still dealing with. Classrooms that have no full-time teacher. So I think what we have is this contrast, which is back to school, is this incredible flood of new and innovative teaching and learning and educator possibility.

And at the same time, the administrative layer is in quicksand with all of the kind of challenges that we've been covering over the last few years, and they're not going away. So there's two schools of thought. One is this the breaking point of the system, and there are many who have just developed a longstanding skepticism about the public school infrastructure as it is and its ability to adapt.

Is this the moment that new models flourish or is this the burning platform that actually enables the system to finally make changes? From that industrialized model to a more learner-centered model. That's really what's at stake here. And it's the speed is the key here. It's a 23, 24, 24, 25. We're gonna have some answers here because it's coming to a head.

[00:33:06] Alex Sarlin: It is. As I hear you talk about this, something jumps into my mind that, you know, I think we've talked about a little bit on the pod, but it really is very clear now, which is that education, I think in the past, used to be behind, you'd think of it, it was behind in technology. Totally. You'd have these ridiculous carts.

You'd cart around with your, your laptop. I had one in my class. Yes, exactly. I mean, yeah, they're still, 

[00:33:25] Ben Kornell: they're still around. I had to, I had to teach other educators at the school how to turn on their computer. They didn't know 

[00:33:31] Alex Sarlin: exactly. People didn't know how to do virtually anything. And there was just this feeling of like, we always have two free computers, we always have two slow in internet.

We always have this. We always have that. It was always behind. And if you think about the things that are coming out of this report, cybersecurity, huge one, AI and the need for clear AI policies. Only 2% of respondents said they actually have AI initiatives underway so far, 2%. So it's coming, but nobody knows what to do with it.

We also saw a huge rise in home access connectivity needs huge rise since last year. Going from 30% of people saying we need to make sure our students are connected at home. And you know, we've talked about this forever, but it's coming right now. These issues, cybersecurity, ai, and connectivity at all times.

You know, this is not an education specific set of things. This is the same things businesses are dealing with. It's the same thing consumer apps are dealing with. I think that's really good for the field. Yes, it's coming to a head in that cybersecurity. There've been these huge attacks and we really need to get on top of it.

But what I think is good about it is that when the education system has the same set of general needs around tech, as other sectors do well, there's lots of solutions that other sectors have, you know, glommed onto that education traditionally hasn't, 'cause it hasn't just even been looking in this direction, but things about interoperability, cybersecurity solutions, ways to improve, you know, digital broadband and access at all times, and mobile access.

Ai, obviously everybody in the whole world is trying to figure out the AI strategy. The fact that education is getting closer to the sort of norm on this, I think is very, very good for the field because it means that education can evolve at the speed of tech instead of just this 10 year lag. Like when you go to, you know, countries and you see the movies that you be the, the movies and the theaters were like three years behind the movies that were in the US and now that's really not true anymore.

Everything launches at the same time. I think this is good 'cause it puts us all on the same page and it allows educators and education, technology directors to actually learn from other fields rather than learning from each other only and always being behind. You know what I mean? 

[00:35:39] Ben Kornell: It is both like one of the biggest conundrums we've had and one of our biggest opportunities I do.

This is also why education is quite interesting. The complexity of it allows us to be leading in some areas or spaces and lagging in others. And I think where we are seeing the kind of leaning in is really around teacher efficacy tools in K 12 and in higher ed, it's around student driven studying tools.

But where we're lagging in K 12 is around the student personalized support. And where we're lagging in higher ed is probably more around the instructional models. So it's very, very nuanced. But I will say, you know, the idea that the best. School of 50 years ago will look exactly the same as the best school 10 years from now.

The odds are very, very low on that. And that has not been the case for a long, long time. I don't know how many decks I've seen where they show like a black and white photo of the old school and a current color photo and it's the exact same thing. Well, that ain't happening anymore. So that's cause for celebration.

Speaking of higher ed, let's talk a little bit about how things are shaking up in higher ed. And you know, just to kind of recap for our listeners, we've been talking a lot about the ROI in college and questions that have really come to a head around, you know, is the four year degree worth it. And then we've also talked about funding changes, student loan forgiveness, and now the focus is really around admissions.

So you know, you're our higher ed guy. Tell us a little bit about how the admissions world is transforming. 

[00:37:20] Alex Sarlin: Well, there was a really interesting survey. Again, this is survey season of admissions officers reported in inside higher ed this week that had some really interesting and, and actually somewhat internally paradoxical findings, which was really interesting.

So some of the things that I think stand out as just very clear are test optional is becoming. Absolutely the norm. I mean, we kind of knew that, but this is now like for sure, only 3% of respondents to this survey said their institution still requires standardized test scores. 3% and almost half, 44% said that their institutions recently switched to test optional during the pandemic.

So you saw a whole bunch of schools that were test optional before and virtually all the rest switched to test optional during the pandemic. That is a big change and that is something, you know, I don't know. I never saw it in my lifetime until the last few years, but it suddenly, it's like. Completely swapped from where it always was.

It was always S-A-T-A-C-T. That's what, that's what it means to get into college. Now they're just more and more making that optional. It doesn't mean it doesn't matter. It still does matter. They still do look at it, but it is optional and it is sort of not the main thing. That's cool and interesting.

You're also seeing this stuff about legacies, which I think is getting pretty clear as well. So as you know, has all these admissions discussions changed in the wake of the Supreme Court decision? Legacy admissions have been under the spotlight because they're basically affirmative action for people who are children of alumni or you know, have various types of legacies.

What you're seeing now is that. Of the admissions officers surveyed 50%. 50% disagreed with the statement that legacy applicants should be favored. Right? So a full half say that legacy applicants should not get preference, and then almost 40% say they neither agree nor disagree with this. They're neutral on it and only about 12%, like one out of eight of these admissions officers are saying, yeah, legacy should get preference.

One out of eight are saying that legacy admissions should get preference that has been baked into higher ed system in this country. The whole time, hundreds of years, and yet now only an eighth of admission officers are saying they believed in legacy admissions. That is a crazy change in a very short time and I'm hoping that it basically, for alls the end of legacy admissions, I mean, you know, it's, it is very clear that it may be going this way.

This is one of the most inequity causing cycles in all of America. And finally, people seem to have totally woken up to it, including exactly the people to whom it really matters. Admissions officers, we haven't gotten to the paradox yet, but Ben, I want to get your take on both of these. What do you think about test optional or legacy?

[00:40:06] Ben Kornell: Well, on the test, optional, I'm skeptical because there's often shortcuts to find ways. To categorize and sort. And if you know, the more, in a way, the more transparent the scoring is, even if there's flaws in it, the more it gives a fairer shot to everyone. And then as you start taking away these clear metrics, it actually becomes more complex for families that don't have experience navigating college admissions.

So what I worry about is we're trading one flawed system for another flawed system. You know, essentially it is a vague and hard to nail down system, which means that someone's going to read in your resume, that you submit, they're gonna infer things about you based on how you present that. And as we all know, like resumes in the job process are all, you know, that evaluative process is very flawed and subject to quite a bit of bias.

Yeah. On the admissions officer side, the one out of eight. Feeling that legacy shouldn't be true. I bet you should be true. Should be true. Only one out of eight feel that it should be true, meaning seven out of eight say we shouldn't use legacy admissions. I'd premise that that was probably the sentiment for many, many years.

It's true. And that that it's not the admissions officers who really make the call on this one, it's gonna be the president of the university. And that's funny. You're right. So I really do think that they're asking the wrong people because if somebody writes a million dollar check or names a building, is their kid getting in?

That's the question. This is an economics question. And the economics question threatens. The colleges themselves and their existential existence because it's these big dollar donors that actually paved the way for a fair amount of the university bloat. And so I do think, you know, I wonder, okay, so legacy admissions is out.

How about prepaid, like buying your kid admissions because. Is that out too? Because if it is out, then it's equitable, but I don't think that's out. And in fact, I think it doesn't really matter whether you are a legacy or not. Like if you give in the millions, is your kid getting in? Probably. And on top of that, I think the, and you may be going to this with the paradox, there is something to be said for, and I'm taking the other side of this, having myself not been in a legacy admission and having gone to school with many other legacy admissions being like, what the F?

But right on this front I will say, I. Going to school with people of privilege when you come from not privilege has an impact and an effect on your network that accelerates your opportunities for success. And I think part of the paradox, and you may be talking about this in a moment, is when you have really wealthy families going to school with kids from low income and middle income.

One, it helps the low income and middle income families access the middle class and accelerate their growth trajectory for their family. So on an individualistic basis, it's optimized, but two, it helps those wealthier families get outside their bubble and understand a little bit more beyond their own circumstance.

So I think this is a complicated question, and I think looking at it with this binary lens really limits the kind of appreciation of the nuance. 

[00:43:38] Alex Sarlin: Great points all around. And there are benefits in having people exposed to others who have very different backgrounds. That said, I am a pretty, I don't know, maybe it's just my nature.

I am pretty down on legacy admissions if they were to change it, nor like by 90% you'd still have a lot of diverse socioeconomic status for me, for me to defend legacy 

[00:44:00] Ben Kornell: admissions. I mean, I remember for years I would tell people, my freshman roommate was Grace Malay, Provo Murphy ii. And you know, I like gave my dad a hug and he was like, gave a handshake.

Goodbye, sir. And that was, that was my introduction to like higher ed. But I, yeah, exactly. But I, I do think we're naive if we think that some admissions policy is gonna. Really Im, you know, implement systemic change. I don't think that's true. 

[00:44:28] Alex Sarlin: You're definitely right. And that's a fantastic point, that it's not the admissions officers who truly, you know, decide this and that.

They probably have been letting people in while holding their nose sometimes, you know, for legacies, for generations. I think that is a terrific point. I am still optimistic. I think that the spotlight, being on this in such a big way is going to create a level of cognitive dissonance that will create some kind of change.

We have seen, already seen some schools just eliminate them entirely. Especially liberal arts. Small liberal arts schools. But we'll see. We'll see where it lands. But I hear you. I mean, the things you're saying make total sense and actually the paradox, there's multiple paradoxes here. One is yes, there's this incredible tension between we want a diverse campus that includes socioeconomic diversity.

That means, you know, we want people from different backgrounds. That includes legacies. The other one, and it's totally related to your point about economics. Schools are running out of applicants. And I mean enrollments, we've talked about enrollments being down for a while, but this is, you know, 220 something admissions officers.

And one, the paradox here is that the majority of respondents said they have failed to meet their enrollment goals. So more than half of the admissions officers surveyed saying we didn't get as many students as we wanted to. And that's been true for a couple years. And it's, people are expecting it to be true in the future.

'cause demographics have shifted a lot, so they don't have enough students. I mean, Ivys and the elites, Ivy Plus aside, they don't have enough students. At the same time, these same admissions officers are saying 80% of them say they did not expect their institution to enroll fewer students in the coming years.

That is the paradox. Somehow these admissions officers are seeing the landscape saying, yep, we're not able to get as many students as we want. We just, there's lots of reasons for that. Yet, at the same time, we will be fine. We'll, we're not gonna have to change anything. We're not gonna have to shrink our classes or do anything about it.

That is a very strange stance to me, and I think one of those stances is not gonna prevail, and I'm pretty sure I know which one it is. Schools are going to have smaller classes, there's gonna be more competition. Colleges are gonna close. And it's really interesting to me that there's this sort of denial is prevalent in this admission officer, you know, cohort.

They're like, yeah, we happen to not have found enough students, but that's probably a passing phase. I don't, we'll be fine. Mm-hmm. Not gonna happen. And that is gonna put even more economic pressure as you're saying on the, you know, high. High. Well, and also forces you to lower 

[00:46:58] Ben Kornell: your standards to have more students admitted have to.

And then the line you were holding on legacy admissions fades away to irrelevance because you start admitting everybody. And by the way, on the flip side, high school students are encouraged to apply to 20 plus schools now. So you're also just seeing, you know, a scattershot approach where good luck getting that legacy student to choose your school because they're gonna be applying to 20 other places and have a lot of different options.

Different options that where everyone is desperate for the enrollment. So this is the part of the iceberg that's above the water. There is so much below the water. And these schools are these large Titanics heading directly for these icebergs? I 

[00:47:45] Alex Sarlin: know, I feel so bad 'cause I feel like we talk, uh, smack about higher ed a lot or I do, you know, and almost every week.

But man, I just think this is a part of the, you know, the system, part of the system in the US that has been cruising along for so long and is now. Headed right for this iceberg, this demographic iceberg. Yes, this perception iceberg. And they just do not know what to do. Grab 

[00:48:09] Ben Kornell: the lifeboats. Get off the ship.

Get off the ship. Well, speaking of grab the lifeboat moments. And by the way, I will just say both Alex and I, when we take a critical stance, we also know there are so many great people, of course, amazing people leading these organizations. Of course driving the work on the ground. And so with that, it pains us to kind of come to the world of Baju where smoke became fire, became a blaze, and now here they are.

In the fire sale moment trying to sell off assets to pay their debt. It's a really sad evolution and development and also a revelation that they hid $533 million in a hedge fund once run from a Miami pancake shop. We're starting to go into Netflix territory where really we're just, you know who's gonna write the script of this, because it is going from good to bad, to worse, to provocative.

[00:49:13] Alex Sarlin: Too absurdist. Yeah. To to Adam McKay. Yeah. This was such, and it's not just any Miami Pancake shop. This is an ihop, of course, this is an IHOP in Miami. But I mean, the Pancake Shop aside, just what keeps coming out about pie juices is that they're just doing incredibly strange things with money. And we know that.

That's why the board of directors, basically half of them walked out saying they don't take our advice about money. I say this every week, but $500 million, I mean, every number associated with bi Jews is a hundred times bigger than every number associated with any other EdTech company. And it's just so absurd how this is going.

And it feels like the big short, and it feels like obscene amounts of money flowing in crazy directions. Or like the GameStop movie that's coming out and it's starting to really crack at the seams. 

[00:50:03] Ben Kornell: Yeah, so let's talk a little bit about what is going on, and maybe I can explain why it might be going on.

So there's basically $533 million and by juice transferred that 533 million in cash to a hedge fund in Florida. And the total amount of the debt is $1.2 billion loan. Now here's actually a little bit of 1 0 1 of how venture debt works. You would think venture debt, they would say, here's 1.2 billion, and you spend down your cash and then you start pulling from that money and it basically keeps you afloat as you're moving forward.

And in some situations it does work that way. Often the collateral for the loan, basically what they can seize if you don't pay it back, can be purchase orders for future money. And the best use of debt in like a startup situation is for cashflow purposes because you're kind of borrowing. Just to keep, you know, the money coming in knowing that you've got new revenue coming in.

Maybe it's a seasonal business or something like that. Venture debt, on the other hand, and we don't know if this is pure debt or venture debt, one, they'll take a little bit of equity of the company in order to give you the loan, but also they have covenants, which basically means you've gotta keep a certain amount of liquid assets on your balance sheet.

So that $1.2 billion loan actually is more like a six or $700 million loan. 'cause you've gotta keep $500 million in cash at all times. So probably what happened, and this is speculation, is by juice, is like, okay, we're gonna keep that money, but we're gonna put it into. A different fund so that the creditors can't claw it back.

The moment we miss our payment, which they did, they might also put it in a fund because they're like, we want that 533 million instead of sitting here to earn some interest. Which by the way, a hedge fund is a super risky place to put $533 million. What doesn't look good is that there was a 2023 Ferrari Roma, a 2020 Lamborghini Huan, and a 2014 Rolls Royce all registered to the 23-year-old running said fund out of an IHOP in Florida.

And this is the shake your head moment where I'm sure the board found out about this and they're like, really? Okay, come on. And so there's like a logic of why one might to delay a bankruptcy, take that cash and put it somewhere out of the creditor's reach, but then you know, they should expect a lawsuit.

It's just the decision making of where they chose to put it. According to these reports. We don't really know for sure that this is accurate, but we've been saying that for basically a year and a half and then it's all turned out to be true. Yeah, I know. And so now here we are, buy juice really was once the most valuable startup period in India, not ed tech startup.

Period. And number two, most valuable EdTech company in the history of the world. Now they're doing a fire sale on some of their best assets, including Epic, which is like, you know, it's a super awesome product and they're gonna end up having to sell it for like 80 cents on the dollar because they need the cash to pay off this big loan.

So ultimately, really, you know, I think the story here is around well-intentioned founders. You know, Baju started as an educator. They haven't been listening to the advice that they've been getting at a minimum on the PR front, and now they're really in a world of trouble. And you know, we're cheering for them to unravel.

Probably also looks bad that they did the whole Lionel Messy sponsorship at the World Cup because it would've been really nice to have that money back right now. 

[00:54:02] Alex Sarlin: Yeah. I mean, the whole thing feels like, I don't even know what the metaphor is, just somebody who stumbles into an absurd amount of money and just does everything they wish they could ever do, right?

They get the most famous soccer player in the world as a spokesperson. They buy every company they think is doing really cool stuff. Who knows? I'm hoping as always that. What this doesn't mean is that Indian ed tech or global ed tech or Asian ed tech or however you wanna slice it, is not, you know, crippled by this absurd story.

The same way that, you know, female founders were kind of knocked sideways by the Theranos story. It's like when people have these biases towards the, the things that stand out the most, and if you hear Indian Ed tech and think bi Jews and think Biju is, did all this stuff, then what does it mean for Physics Walla and up grad and TU and all of these companies that are really trying to do amazing things, A LEAP school.

So I'm hoping that, you know, wherever Bi Jews lands, we can sort of separate out this story as a absurd outlier from the EdTech world. 

[00:55:05] Ben Kornell: Well, and you know what, probably Alex, at some point there were people in the ear of the founders saying, grow, baby, grow, acquire m and a, raise more money. You know, all of that kind of stuff.

And I think this also just shows that the hangover. From the 20 20, 21 peak can be very severe. We haven't even covered paper in today's show, but it's another example of where, you know, kind of the market potential really didn't pan out, and now there's severe consequences for the company. Let's go to wrap up with some funding, mergers, and acquisition.

First one up I had learn X, Y, Z securing a $3 million seed round. And their product is now posted on Product Hunt. By the way, Christian and aren't are two of my like absolute favorite founders. They have so much joy and Christian was at LinkedIn. Basically the world is his oyster and he's like, I want to use generative ai.

I. To create dynamic, responsive learning opportunities for anyone, and it's just a great product. And they raised 3 million. Really, really exciting. 

[00:56:18] Alex Sarlin: We also just added them to our lineup at the conference in October, Christian will be speaking about personalized learning, so that's really cool. 

[00:56:25] Ben Kornell: That's great.

And then we also have Cure Pod, one of our companies supporting, you know, educators with generative ai. They basically help teachers make interactive lessons. They raise 4.6 million in seed funding. So far, over 150,000 teachers have used Cure Pod to create over 240,000 lessons reaching over a million students.

Wow. That again, the adoption is incredible. What are some that you're looking at? 

[00:56:54] Alex Sarlin: So we are continuing to see AI based companies gather fundings. This week we saw Math Espresso, which uses AI for math, get $8 million out of South Korea. They've had Google among others as investors. We saw Elsa, which is an English language learning company.

It's sort of with grammar and various things that it has an office in SF in Ho Chi Minh and Lisbon, Portugal get $23 million series C round, so that's another one of these language learning apps. We saw a Danish language learning app called swap language. Receive 2 million and a Spanish language and writing app out of Madrid raised 6.5 million euros.

So a whole slew of additional AI based investments including several language tools. Interesting week. 

[00:57:41] Ben Kornell: Alright everybody, we've got our first interview coming up and we've got Viv Ram Krishnan joining us from Project Read, project Read brand new thing, incubated out of Stanford, GSB, close to my heart. And you're out piloting with schools, leveraging ai, transforming reading.

Viv, just tell us a little bit about the company and your journey to kind of get it started. 

[00:58:02] Vivek Ramakrishnan: Absolutely. Thanks for having me on, Ben. You know, project read has taken several iterations, but where we have landed is going all in on building for teachers who are delivering science and reading line instruction In the fur earliest version of the company, we were, we were really focusing on buildings, personalized content and stories for kids as they dream it up using, you know, a suite of generative AI tools.

But what we consistently heard from teachers is that, hey, our district or our school is moving towards, you know, more direct instruction around systematic, explicit phonics science of reading aligned instruction. And we are struggling for a number of reasons to implement it. One of which is we have a shortage of decodable texts.

Second of which is just the cognitive lift on us as teachers is, is higher when we're, when we're systematically teaching phonics. And I have to try to track all of the misconceptions of my students who are in vastly different places as, as far as what their needs are. And we said. We think that the emergence of some of these generative AI capacities is dovetailing really nicely with the need we hear that these teachers are expressing to us, and I think we can build something here.

And so starting in in the spring, late spring, early summer, we really went all in on building four teachers and their students in the classroom. And we piloted with a couple of schools over the summer and then we released our, our first product, our AI tutor, just a couple weeks ago. Uh, and the uptake has been awesome.

And then just on Friday we released a decodable generator for teachers, which we built in response to just an outpouring of feedback. We heard from them around their need for more decodables. 

[00:59:33] Alex Sarlin: So, you know, we just saw news this week that Teachers College Columbia is sort of beginning to break ties with Lucy Calkins and, and there's, it's sort of becoming very, very, very official.

We have more than 40 states moving into this science of reading world. You've been, you know, upfront talking to all these teachers, making this transition and you mentioned a couple of things they're struggling with, but I'd love to hear just your take on this entire moment when it comes to reading in the us.

That's a great question, Alex. 

[01:00:01] Vivek Ramakrishnan: You know, it's interesting because I think the conversation at the district level, at the state level, legislatures have acted on this and enacting, you know, laws around what sort of curricula that schools will need to employ. And there's a grace period in many states, but that's where the, the puck is heading if it's not there already.

Now with that being said, when you actually talk to educators, there's this mismatch between. Where we're going and how ready they feel to tackle that. Mm. Because so many of them have been traditionally trained in, in balanced literacy methods, most prominently three queuing where, you know, you ex you give a kid some generic hints around, Hey, what's the first letter?

What does the picture suggest? Does that make sense? And I recall, so my background is, is in education and in schools. And I stepped in as our first grade teacher at the charter that I found and was running before grad school. And I just, you know, I was blown away by the cognitive lift, both on me to, to learn how to teach phonics well.

But also as I mentioned, all of the considerations for each kid, I now had to juggle in intervening in the right way for the right, at the right time for that student and giving them enough text to, to support that. And so it's no surprise to me that individual teachers, even if they support the direction of the shift, that they feel overwhelmed.

And I think that's really what we're trying to address. Yeah. 

[01:01:24] Ben Kornell: Yeah. I love that you're pairing the technical change and the adaptive change together to really support them there. Um, your background is so fascinating. I'm curious, how is founding an EdTech company similar or different to founding a charter school?

And what are some of the lessons that you've learned that have been transferable, and what are some things that you've had to relearn or, or kind of start from scratch? 

[01:01:47] Vivek Ramakrishnan: Great question, Ben. In many ways, starting a charter school is truly, you know, it, it is building a startup. There's just no exit potential.

There's, there's, there's more regulations you have to deal with. So in many ways it actually feels like there's way more variables that we have to solve for, as well as just kind of. The human responsibility that comes with serving staff, serving families, and, and balancing what that looks like day in and day out.

Not that that's not the case in EdTech, but at least for a small company like us, we're really just focused a lot more on product and engaging with external stakeholders as opposed to internal stakeholders. That's pretty different learning curve for me. Certainly on, on the tech side, you know, my background is, is not within tech and I took a couple compsci classes at Stanford, but you know, one of the, the best ways for me to turn that to accelerate my, my progress on that curve was working with John Danner, who I was fortunate enough to meet at Stanford and we wrote the MVP.

John is still a great backend engineer and I did what I could to be serviceable on the front end, but actually being in the code base, just the two of us for months, getting really deep into both the use of the AI tools, but also figuring out like what it means to build an early stage product. And I think as we have brought on, you know, an engineer now full time.

My ability to see around corners. Uh, I'm still learning every day, but it's very different from if I think we just like outsourced engineering from the beginning. 

[01:03:15] Ben Kornell: Yeah, I love that. And it's also a great message to founders to like, don't be afraid to get in the code and just, you know, there, there's so much you're gonna screw up and so much that you're gonna learn through doing, making mistakes or doing the wrong way.

But it's almost that experience that allows you to be a better leader for future subsequent engineering teams who are gonna do it way better or way more efficiently. But that visceral experience of like building the MVP, uh, just has a ton of value. 

[01:03:45] Vivek Ramakrishnan: Well, you know, Ben, I think that actually brings up for me what the biggest parallel between building the charter in the early stages was.

And this is just wearing a bunch of different hats and you what each day will. So we were so early back then, and now, you know, we're at a different scale with the, with the charter, but I was our chef for three months every day. Our main chef when we had some turnover in the kitchen thing for two months when we had a custodian who had a back injury, you know, stepped in as I mentioned, as first grade teacher.

And my ability to then scale those various functions of the organization was so much better because even if imperfectly or for short periods of time, I had to wear those hats. And I just love the versatility that, that the day brought, you know, early stage startup life, whether it was building a school.

And I think the same thing is true right now in building project read. One day I'm trying to figure out how to, you know, manage our SQL queries and, and learning SQL the next day. You know, we're, we're looking at conversion rates same day. I'm talking to teachers about what's working in the classroom and what improvements they want.

And it's just really fun, I think at this stage across the sort of product, whether it's a school or an ed tech, to wear a ton of different hats. 

[01:04:51] Alex Sarlin: Yeah, it sounds like that you learned a lot being the chef and the, you know, custodian. It's really, it's very clear example of, of having to lean in on, you know, areas that you might not be an expert on.

Definitely true in both environments. So, I wanna ask a little bit about your use of ai. We, we, you know, you, you mentioned this AI tutor. Can you just give us a little bit of a run through about what the experience is like for, for a teacher and kid using Project Read as of right now? I know you've just been launching now.

[01:05:19] Vivek Ramakrishnan: Yeah. So as it currently stands, teacher within 90 seconds can log on a, create a profile, pick their school, and set up a classroom, and they can set the starting objective according to our scope and sequence for that classroom. Different teachers will start the classroom at different places depending on, you know, what grade they're teaching, whether it's an intervention group, so on, so forth.

But once they set up their classroom, students can log in with a code and quickly start reading to the tutor. And what happens, a couple things are dif are happening as a student reads that decodable text at the phon level, they're getting real time feedback from our ai, Hey, you did an awesome job, but can you try that word again?

They try the word again. Then they get a phon specific hint, you know, the short a sound like in the letter or like in the word Matt, you know, and so on, so forth. And they're getting specific feedback based on the errors that they're making. And the other thing that's happening is we're logging words that they miss and incorporating those into future stories.

So, so stories are being generated and personalized for each student, not just based on the phonics skill they're working on, or the mastery they've shown, but also what gaps that they still have until we see that they don't have those same gaps anymore. And we're currently working on, we have a pretty simple dashboard right now for teachers, but one of the things we've also really been in the weeds talking to teachers about is like, what sort of data and how could that be presented that would actually be most helpful to you as you are trying to take all of this stuff that we're getting from the AI and fashion further instruction in your classroom.

So that could be most commonly missed, irregular words that could be phoning graphene pairs that each student is struggling with, has mastered trends, the whole class level, so on so forth. So, you know, dashboards that ad techs have built for years, but really specific to the sort of phoning level data that we're getting.

[01:07:01] Ben Kornell: What I love about your company is that in some ways, two or three years ago, it would've been impossible to build. What you have. So it's really at the nexus of cutting edge technology and the potential of ai. And two or three years ago, the science of reading was not in a space that's true. Where it, you know, it could really take off like this.

And, you know, so much of, uh, the studies actually show that for startups to be successful, the number one factor is timing. And so there's a really, I, I mean, I just think that that's one of the most exciting parts about this moment that Alex and I get to cover is just seeing all of these ideas that have many of us have thought about for a long, long time.

And, you know, you're putting it together in ways that are just super inspiring. Okay, last question. I'm the business guy. Yeah. John Danner, uh, for those of you who don't know, is like a legend in EdTech. He's an angel investor through DUNS Capital. He started Rocketship, uh, charter Network. He also has been an incredibly successful startup, uh, founder and so on.

But he, he has these like mantras or kind of like a Bible of ed tech and like, number one thing in the Bible is don't sell to schools. And, and then number two is make it free. And so I'm really curious. You're going, you, you know, it's probably the most shocking development of the whole evolution of project read is that it's actually focused on teachers in schools.

Thinking about how to empower them to unlock things for learners is the antithesis of many of his playbooks. And I just love to hear how you landed on that and also what is the business model? How does this thing scale? 

[01:08:46] Vivek Ramakrishnan: Yeah, good questions. All the above. Just to back up, I think, Ben, you might have even been aware of some of the previous iterations where we really did test this with a diverse set of parents.

Yeah. And what we consistently found is across the board, students loved what we built, but the parents that kept coming back to this, those that you know, were willing to sign up for a subscription for 20 bucks a month or whatever that would look like. Were not necessarily those parents and guardians of students that we really wanted to move the needle for.

And so I think John and I kind of had a reckoning pretty early on where we were, I mean there wasn't really even a question, but like we're hearing this need from teachers. We know the students we really want to focus on in reach. I. Mission's gonna win out. And so we have not yet venture capitalized this.

And maybe at some point that happens, but we haven't gotten there yet. And in, in some ways, I think really my hope here is that this is an organic teacher adopted movement. I mean, we've had both charter networks, districts pilot, stuff like that, but we've made it free. Two teachers for the 2324 school year just to one, continue to iterate on the product, but also spread the reach early.

And then hopefully that's kind of a pathway to reaching more schools when we do wanna ultimately monetize. It's an interesting question, it's a hypothesis. I think we're still proving out. You know, traditionally I think a product like this would go with sales and our approach here. Has really been to try to find, in some ways, very true to John Danner, what are the, like the viral loops that would bootstrap adoption at the teacher level.

And that's really where our focus has been, that we have engaged in parallel with, uh, principals, chief academic officers, district stakeholders. So those are kind of two parallel hypotheses that we're running. But I think besides the business model for my, my personal theory of change is that if we can get this at scale in classrooms, working for teachers like that is how we're gonna reach the most kids possible as quickly as possible, independent of business model.

[01:10:47] Ben Kornell: Amen. Yeah, I, I think anyone who believes that a B2C direct paid model is going to deliver the kind of equity and impact that they're looking for is, is not being honest with themselves. And by the way, I'll also say to our listeners. I love selling to schools because the lifetime value and the recurring revenue is so much better in schools like consumers churn schools once they find something that they love.

I mean, you can go to a school now and they're running software from like 2013 because they just love the software. That lifetime value is really great. So I love this idea of finding the viral loop with educators who are experiencing a profound pedagogical shift, enabling technology that has profoundly shifted.

And then ultimately, you know, figuring out where are the opportunities to monetize, either in whole district or upselling. All of that fits so well with your mission. Viv, what a joyful pleasure, and I hope you're having some gourmet deluxe ramen as you're building this thing. And it's so inspiring to bring your story to our listeners.

Thanks so much for joining us. Um, if people wanna find out more about Project Reed, where should they go? 

[01:12:04] Vivek Ramakrishnan: Www dot project read ai. 

[01:12:08] Ben Kornell: Wonderful. All right. Thanks so much Viv. 

[01:12:10] Vivek Ramakrishnan: Thank you so much for having me on. Appreciate it guys. 

[01:12:13] Ben Kornell: Alright. EdTech Insider listeners. We are joined by a great colleague, friend and EdTech, CEO, dup Pancho Aurora of Todd.

I have to tell you the moment that I demoed Tottle, I was like, oh my gosh, this is next level the, these are things that I had been working on my entire career and Tottle had really unlocked elements of planning, learning, assessment that were just it. It was like seeing into the future when I first demoed.

So I'm so honored to have you, Alex and I are so honored to have you on the show. DuPont you, Aurora, welcome. Thank you. Thank you so much for having me here, Ben. Well, we have so much to talk about, but first, let's just start with the founding story of Tattle. Where did Tattle come from and tell us a little bit about the product itself.

[01:13:02] Deepanshu Arora: Great. So I'll just give you a backstory first. Uh, before founding tole, we as a team founded a network of preschools in India and we also run a progressive K 12, uh, IB World School here, here in India. And as a part of our school and preschools, we were looking to integrate technology meaningfully. I've always had a very teacher's first lens to technology integration in classrooms with the belief that till the time technology isn't solving for a teacher's pain point, it's not solving for a teacher's workflow.

Everything else is kind of, uh, meaningless. So we were looking to integrate technology in our own classrooms as we went down that journey. What we realized was that to manage our own teaching and learning workflows, we had to use four different platforms. So we were using one platform for curriculum design, another platform for doing student portfolios, a third platform for doing assessment and reporting, and a fourth one for doing family communication.

And it was a very messy work experience, uh, very messy user experience for our teachers to straddle between so many different, uh, platforms. So we thought to ourselves, why don't we build something that will simplify, uh, the workflow for our teachers? And we started building it. We never really had it in mind that we will be building it as a company and taking it to schools around the world.

But, uh, serendipity happened and uh, we were hosting some conferences at our school. As a part of that conference. Uh, we had educators from different parts of the world participating. They had a look at the platform. They really liked it. They said, guys, why don't you build this in a scalable way? And that's how we decided to take a plunge and.

It's been a very rewarding journey. Today, TOLE is used by about 1500 schools from different parts, uh, uh, parts of the world. Uh, at the core of it, it's a teacher workflow platform specifically designed for progressive schools. So on tole, teachers can do their curriculum design, which includes building out their whole school curriculum maps, creating their yearly learning plans, designing the unit plans, lesson plans, assessments.

Then they can assign work to students through total, collect back work from them, give them feedback. All the assessment data for a particular student keeps getting added to the student, uh, profile. And whenever teachers want that student profile also converts into progress reports, which can then be shared with the parents.

So in a nutshell, that's the core workflow of the platform. If you think about it, that's also the core workflow of a teacher. So we are trying to bring all of that together meaningfully in one place. 

[01:15:23] Alex Sarlin: It's amazing and looking at some of the coverage that you have. My background is in instructional design and it's, I just feel like you are so deep in, you know, I see project-based learning, inquiry-based learning, understanding by design, authentic assessment, social emotional learning, you know, as well as your ib, but you also do, uh, American and British curriculum.

That is a wide lens. You mentioned sort of doing progressive school. These are some of the most impactful and powerful instructional design methods. How do you think about sort of integrating this type of very high level proven instructional design into the workflow for teachers? 

[01:15:59] Deepanshu Arora: So, for us, the way we have designed the platform is that it's fairly malleable and it can be, uh, adapted to a school's, uh, context as long as the school is doing two things.

One is a focus on teacher led curriculum design, and second is competency based assessments. We have seen that if a school is doing those two things, then total is a really good, uh, fit for them. 

[01:16:21] Ben Kornell: Let's talk a little bit about AI and how you've integrated AI into this latest release of total ai. And I will just say, I think there's a lot of companies that have been kind of AI, one trick ponies where it's like, okay, AI can do this one thing, but you're already sitting on this kind of comprehensive teacher workflow solution.

So building AI into that is infinitely more complex. Tell us a little bit about how you've thought about AI and also like what has it been good for and where do you see 

[01:16:54] Deepanshu Arora: the limitations? So our thinking about AI in education is largely guided by two beliefs. One is that we believe as a team, as a company that the biggest impact of AI in education will be in, uh, supercharging teacher work.

Mm-hmm. For a very long time, we have had these, uh, collective notions around what progressive education or what future ready education should look like. We have always been talking about the fact that education needs to be personalized. It needs to be project based, it needs to be connected to the real world.

It needs to focus on skills and competencies rather than just knowledge. And, uh, a lot of times these just end up being buzzwords because the reality is that, uh, we can think about all of these things, but it's the teacher at the end of the day who has to implement this in her, uh, classroom. And teachers as we know, are already super, uh, overworked.

So our thinking really was that how can we, I. Leverage AI to one, simplify teacher work and second, help them elevate their practice. So that's one. And the second, uh, belief for us is that AI will be most powerful if it gets integrated into the teacher workflows rather than it being something that sits on another place.

And then teachers have to keep going to it to use its, uh, to, to use its power. So therefore, we are integrating AI really deep into our platform. For us, the thinking is that AI is, uh, not just a feature, but it's really a core capability of our platform. I. It's the platform itself, and therefore we are integrating AI into all teacher workflows from curriculum planning assessments, progress reports, family communication, data analytics, and really thinking about how can AI, uh, simplify each of these things for teachers.

As a part of our first release of Total ai, which we did, uh, around two months back, we launched Total AI with three use cases. One is lesson planning. What we realized was that teachers spend around 10 hours a week doing lesson plans and a lot of preparation time goes into it. Uh, so we have built total AI to help teachers create really authentic, really high quality lesson plans in a really fast way.

So what took hours earlier can now be done in a matter of minutes with total ai. The second feature that we launched, which is, which seems kind of. Counterintuitive, but teachers really love it is writing comments in student progress reports. So every term teachers spend around 40 to 50 hours writing personalized comments for each student in their progress reports.

And what we realized was that as a platform, we already have a student's assessment data with us. Can we really simplify that process of generating personalized comments for teachers? So, and teachers, absolutely. Uh, love it. I just before this, I was on a call, uh, with a school leader who said that this is going to save their teachers about 25 to 30 hours each term.

And that's like really powerful. So that's the second thing. And the third feature that we have right now launched is a writing assistant. So teachers do a lot of writing. They need to create a lot of resources. It's a simple integration of AI in the platform where anywhere teachers are writing content, they can leverage AI to write that content faster and better.

In the next version, which we'll be launching in the next, uh, two to three months, teachers will be able to design assessments with ai. They'll be able to evaluate student work using ai. They will then we are also building a, a PD assistant, a professional development assistant within total. So if you're creating a lesson plan, can you have AI as really a thought partner in creating that, that lesson plan?

So for us, it's all about teachers. We have always been teachers first in our approach, like I said earlier. So we are really going deep into solving for the, uh, teacher workflows here. 

[01:20:44] Alex Sarlin: I can only imagine how the teachers and administrators and, you know, react to that level of efficiency and time saving, as well as the fact that it's still integrated with all sorts of really best practices for pedagogy.

It's not just, you know, Hey, we'll write you a lesson plan so you don't have to do it. It's, we'll work with you to create a extremely, you know, sound lesson plan that's really gonna help elevate your teaching and it's gonna save you time. We're at this moment where the teaching profession, you know, throughout the world is under a little bit of a stress.

I mean, maybe more than a little bit coming out of the pandemic and just all sorts of things happening in different countries. Do you see this time saving as a way to sort of retain teachers and help the teaching profession become more of a stable and sustainable, you know, profession for anybody? 

[01:21:34] Deepanshu Arora: Yeah, that's a hard one.

I. I don't think that we should see AI as a silver bullet that will solve for everything. I think there are a lot of systemic issues because of which, uh, teachers are leaving the profession, uh, like never before. And AI can yes, solve for, uh, simplifying their workflows. It can, uh, solve for making their work a bit more in interesting and a bit more, bit more challenging.

But till the time we don't address issues like teachers pay, uh, we don't address issues like teachers being given the right kind of professional development, uh, opportunities. I don't think that AI will be able to, will be able to solve for that larger systemic issue. 

[01:22:17] Alex Sarlin: Yeah. Yeah, that's a great answer.

That makes a lot of sense. 

[01:22:20] Ben Kornell: Stepping back, you are really a international first company where really you've built a product with the intention of being in countries across the world from the get go. But as an Indian EdTech company too, there's been a lot of boom and bust cycle in the Indian EdTech ecosystem.

How have you navigated that and how's that putting pressure on other you and other Indian EdTech entrepreneurs? As kind of the Baiju SGO plays out, but also many of the other kind of headline grabbing layoffs and ups and downs. How have you approached that? How 

[01:22:56] Deepanshu Arora: have you managed it? So, Ben, that's an interesting question.

I think the problem in the Indian EdTech ecosystem is that EdTech often gets equated with online, and, and those companies are the ones that raised the most amount of capital during the COVID boom. And those are, again, the companies that have been impacted the most. Once COVID caught over and everyone realized that, uh, online learning is not as promising at what as what everyone was really talking, talking about.

So for us, we are in a strange place. Some people look at us as an EdTech company. Other people look at us as a pure place SaaS company because our business model is SaaS. Even though we are working in education, our business model in is, is SaaS. So we kind of. Opportunistic about it. So we, we try and place ourselves into the EdTech bucket when it suits us and other times we put ourselves in the SaaS bucket when it works for us.

Uh, but largely I think in the Indian ecosystem specifically, there is a newfound re uh, respect for companies that are trying to build their business the right way. Companies that are actually generating significant revenues, companies that have, uh, good fundamental unit economics, uh, in place. So I would say that for a company like us, which has sound fundamentals, the whole era of 2022 onward has actually been more positive than negatively now get a lot more investor interest than we used to earlier, because, uh, people see that we are building a real business here and solving a really meaningful problem as well.

[01:24:31] Ben Kornell: That's a great point. And I do think that one, being opportunistic, it's great advice for any entrepreneur in thinking about how you frame what you do to, to really help others understand it. But two, many times the headlines paint these broad brush pictures that negate the kind of reality on the ground that is much more nuanced and complex.

And just so our listeners know, you recently closed around, led by Sequoia, India, and there's a way in which kind of surprise, surprise, good business fundamentals is actually good business to invest in. And this shift from pure top line growth, I actually think is a very good shift for SaaS B2B education businesses because there's such good, uh, lifetime value and recurring revenue in some of those partners.

So, uh, very, very interesting. 

[01:25:23] Alex Sarlin: One of the things I find really interesting about Taal is you also use the platform to bring educators from many countries together around learning and teaching and learning. You have learning jams, you have symposiums, thinka, hons, bootcamps, summits, and conferences. For educators who are really interested in this type of teaching, can, can you tell us more about that strategy?

It's really interesting to hear. 

[01:25:47] Deepanshu Arora: Yeah. So for us, it goes back to our teachers' first mission. Uh, and that is what we try to stand for. So we have, uh, not invested, uh, much in paid marketing. So we have never run Facebook ads. We have never run Google ads and things like that. What we have always focused on is, uh, can we deliver value to teachers?

So our marketing focuses on that. Can we create high quality content that will help teachers become better teachers? School leaders become better school leaders? Our product focuses on that, that can we build products which help simplify teacher workflows. Our, uh, school success team, which is our support team, also focuses on that that can be help teachers extract the most value out of the platform.

So that underlying belief or vision that we are teachers first runs across everything. And, uh, the community outreach that we do, all of the events that we organize, whether they are uh, virtual or in person, is kind of guided by that teachers first philosophy. DuPont 

[01:26:50] Ben Kornell: you. It's such a joy. Anytime I get to talk to you, I always feel like I learned something.

I also get like 50 other new questions in mind. So this will be just one of many conversations we have with you on the show. So glad to have you here joining us. Congrats on all your success. And what's the best way for people to find out more about Tattle if they're looking for more information? 

[01:27:11] Deepanshu Arora: So people can just head to our website, todd app.com, and, uh, they'll find all the information there.

Again, really happy to be here, Ben and Alex. Uh, thank you so much for giving me the opportunity, and I would just love to give out a special shout out to Ben, and I hope that you retain this in the version that you, that you release. Ben, I've known Ben for, uh, a year and a half now, and all the conversations that we have had around building a company and especially building a company in education have been super, super insightful.

I almost look up to Ben as a mentor who I keep going to with all the questions that I have. And, uh, Ben, thank you always for all the advice and the time that you, that you spend with me. 

[01:27:54] Ben Kornell: Oh, man, I am so, I am blushing right now. So much it's mutual and thank you so much for saying those kind things and thank you for joining us for the show.

Now the world gets to experience a little bit of the tattle magic. Thank you. Well, that wraps our show. Thank you so much to Viv, to DuPont You. Thank you all. To our listeners, we hope to see you at one of our upcoming events in San Francisco, in New York, online, wherever you may be. And thanks so much for listening to the week in EdTech, because if it happens in EdTech, you'll hear about it here on the weekend.

[01:28:31] Alex Sarlin: Thanks for listening to this episode of EdTech Insiders. If you like the podcast, remember to rate it and share it with others in the EdTech community. For those who want even more, EdTech Insider, subscribe to the Free EdTech Insiders Newsletter on substack.