Edtech Insiders

Week in Edtech 05/29/2024: OpenAI Announces a New Safety Board, Anthropic's LLM Mapping, New Chromebooks with GenAI for Educators, Guild Education Layoffs, Microsoft Partners with Khan Academy on AI Tools and More!

Alex Sarlin and Ben Kornell Season 8

Send us a text

Join Alex Sarlin and Ben Kornell as they explore the most critical developments in the world of education technology this week:

🚨 OpenAI announces a new safety board, strikes a $250M content deal with News Corp, and begins training its next flagship AI model
🚀 Anthropic unveils a new method to map large language models
💻 New AI-enhanced Chromebooks are launched for educators
🤝 Microsoft partners with Khan Academy to develop AI educational tools
📉 Guild Education lays off a quarter of its workforce
📊 An edtech firm seeks $515M in Abu Dhabi’s first 2024 IPO
💡 Praktika raises $35.5M for AI avatar language learning
📈 NCAA agrees to a $2.8B settlement for athletes over NIL rights

Upcoming Event:
AI Arcade Night at ISTE
When: Monday June 24 from 5 PM - 8 PM
Where: 1Up Arcade Bar 1925 Blake St. Denver CO 90202
Register Here

Don’t forget to subscribe to Edtech Insiders for more updates and insights from the forefront of educational technology!

This season of Edtech Insiders is once again brought to you by Tuck Advisors, the M&A firm for Education Entrepreneurs.  Founded by serial entrepreneurs with over 25 years of experience founding, investing in, and selling companies, Tuck believes you deserve M&A advisors who work just as hard as you do.

Alexander Sarlin:

Welcome to Season Eight of Edtech Insiders where we speak to educators, founders, investors, thought leaders and the industry experts who are shaping the global education technology industry. Every week, we bring you the Week in Edtech, important updates from the EdTech field, including news about core technologies and issues that we know will influence the sector like artificial intelligence, extended reality, education, politics, and more. We also conduct an in depth interview with a wide variety of EdTech thought leaders, and bring you insights and conversations from edtech conferences all around the world. Remember to subscribe, follow and tell your Edtech friends about the podcast and to check out the Edtech Insiders substack newsletter. Thanks for being part of the EdTech insiders community. Enjoy the show.

Ben Kornell:

Hello, everybody, its Edtech Insiders. Once again, we're back, Alex and Ben, bring us around the horn of the school year, it's June. We are really excited to like get on the bathing suits and go hang out by the pool. Because there's been a lot going on this fall. And we need we need a break. But we're gonna sprint to the finish here this last month. And we've got a lot going on on the pod and on events. Alex, what's going on with the pod?

Alexander Sarlin:

Yeah. So last week, we released our interview with Shantanu Sinha, the head of Google for Education, General Manager of Google for Education, which was a terrific interview and actually has become one of our most popular of the year already. It was awesome. If you haven't heard that one, I highly recommend it. We've also talked to Mark Heaps who's the sort of evangelist for a new LLM called Groq. You just interviewed Tory Patterson on the podcast from our who is you know, a huge edtech legend. And that won't be coming out soon, as well. And we have a whole series setup for the next couple of months.

Ben Kornell:

Yeah. And then on the event side, we have a fun event. We've never done this before. But workshop ventures and Mike Berlin are hosting an edtech baseball game at the Oakland B's. So the Oakland A's are moving on to Sacramento and eventually Las Vegas, and so on June 12, will be out at the ballgame. And you can pick up tickets, check it out through our newsletter, you can find the link. And then we're also going to be in Denver at SB on June 25. We're teaming up with our friends over at risk learning to do an arcade night where it's basically unlimited video games and unlimited edtech conversation that's on the 25th in Denver, Colorado. We're excited to see you there. Really excited to see you there.

Alexander Sarlin:

Yeah, ISTE is sort of the last of the really, really big EdTech spring conferences. And it's been tiring. It's been a tiring conference season, but also an incredible one. I think everybody is just trying to figure out how to combine the two messages of AI just continues to move at a breakneck pace. And we'll be talking about that today, as always, and at Tech funding is way down. And all the startups are trying to figure out exactly where to turn their chips. It is a crazy moment. So why don't why don't we start with a little bit of a around the world with some of the sort of big AI news. And then let's get into the actual edtech AI news, which is also super interesting this week.

Ben Kornell:

Yeah, so I'll kick it off on open AI, there's been a ton of drama, I feel like there's a Trumpian theme which is like, no publicity is bad publicity, perhaps. In addition to some of the mishaps with Scarlett Johansson, his voice and so on, there's quite a stir around their safety team. And let's just be clear that the safety team that was dismantled was really around like AGI safety and the threat to human existential, you know, existence. But what they really done is they've taken their safety board and combined it in with their core product team. And so they are looking at things like existential risk alongside data privacy and compliance and some of the more mundane pieces. Ilya Sutzkever, Ilya Sutskever, who was famously one of the opening AI founders who voted out Sam Altman in the coup a little bit, almost a year ago, was also part of that departure. So you know, I think what's ending up happening is open AI is trying to figure out how does it brand itself as the safe version, and these kinds of headlines go in the opposite direction? How about and Robin? Yeah,

Alexander Sarlin:

so the flip side of this, you know, open AI, they're trying to create this new safety committee, but some of their lead safety people are leaving and yon like, I'm probably pronouncing that wrong, but let's see if it's right Yan Laika, who was the other sort of head of the safety committee at opening I went and jumped ship to anthropic which is the spin off company that is one of open the eyes big competitors opening I had other news this week, they announced it was in the New York Times that they began training their new flagship model, we could call that, you know, GPT. Five, for lack of a better name, but we don't know exactly what it is. They also made a deal with a Wall Street Journal for content, they started working with Reddit, which they had already made a deal with. So lots of stuff happening there. But anthropic, I think sort of stole some of the headlines today, especially as you say, Ben, in terms of really staking a claim as being the they're really making the claim that they're the safe AI, that they're the one who is actually really being thoughtful about bias and misuse and deception and hallucinations. And Google made some negative news this week for its model, spitting out some strange advice about I believe, using glue on a pizza, which is like came from somebody's ridiculous, you know, internet thread. But anthropic basically said this week that they made a breakthrough in understanding how large language models actually work, how they sort of think internally, this is sort of touches on this whole idea of Explainable AI, which is a big, big and pretty important concept as we move towards that AGI era that you're talking about, then, because we all know large language models, they train on this huge amounts of data and then they sort of piece it together and act you know, very human like, but we don't actually know what they're doing. It's a little bit of a black box. And anthropic is just put out a big paper basically saying, hey, are big, big breakthrough announcement saying I think we can actually start to understand how that works. And then not only did they hire Yan, like from straight out of open AI, they hired Mike Krieger as their new Chief Product Officer, that name might sound familiar, he was one of the two founders of Instagram, which was famously, you know, one of the most valuable acquisitions in all of tech history, and Instagram, arguably one of the most successful products in all of tech history. So that is another big name hire. We got to keep our eye on anthropic, we're gonna see what's next. What about Google? There? Yeah, we can't leave Google out of this mix. They did some really interesting stuff that actually really touches more directly on education.

Ben Kornell:

Yeah, I think both Google and Microsoft are making big plays in the education space. For Google, they had a flurry of announcements. And as you mentioned, at the top Shantanu really walk through some of the use cases. But this idea that it's going to be aI integrated into everything is pretty profound when you're talking about Google's reach. And so they announced new Chromebooks with generative AI for educators embedded. Now this might not seem like an aggressive play, it might seem more like a retention play. But actually, most educators use Apple products for their laptops, while the students use Chromebooks for their products. So it is really interesting that this is a shot across the bow for Apple Google fight. And this also moves them more solidly into the enterprise play where their primary customer is educators and administrators, that really signals a double down on education and education use cases like that Microsoft, they're kind of jumping in on the Khan Academy train, similar to open AI, but their Khan Academy play is free for schools. So very interesting to see open AI, which is really positioning themselves with chat GBT, at least on the front end as more of a consumer accessibility tool for tutoring with your kids learning alongside your children. You know, parent is audience and consumer is the adoption cycle, Microsoft taking that exact same playbook and just saying, Okay, we're gonna drive this on the enterprise side, we're all wondering, when is Khan Academy going to be available for everyone? And, you know, basically, they flip the switch. And now Cognito is free for teachers. And I think it's only in the US is what I understand. But basically, this levels up to the integration with Microsoft co pilot, and teams for education. So big enterprise moves from both Google and Microsoft Enterprise

Alexander Sarlin:

moves, specifically aimed at the educator market, which is really, really interesting. And you know, I mean, those of us who have been in a classroom in kind of in the last few years, Chromebooks are ubiquitous, Khan is relatively ubiquitous, it's certainly used a lot. It's certainly sort of the go to platform for many different use cases in schools. And you can sort of see these camps lining up. I mean, as you're saying, Ben, this Microsoft, we know is very much embedded with open AI, Microsoft and open AI. Really, you can consider them core partners. I think, at this point, Khan Academy, as we announced on this podcast, you know, many months ago, they had early access to chat UBT, 3.5, and then to four, and they were able to really get ahead of things for a very specific reason, because the open AI people wanted a trusted partner in education, somebody that people really love and trust and just like you're saying, Microsoft is now following that same playbook and Microsoft con and open AI sort of together make This triangle on there providing access for conmigo for teachers. So it really is basically the equivalent of the sort of AI for educators on Gemini, which is on Chromebooks. We didn't mention one other thing about Google. And I'd love to hear your take on this. Google announced something called Learn LLM. And these are large language models, specifically trained for the education use case trained in partnership with educators and pedagogy experts to try to optimize for education rather than general queries. What did you make of that announcement? That was interesting?

Ben Kornell:

Well, it's interesting to me, mainly because there's a potential developer play here where learn ln becomes the primary LLN that developers third party edtech tools, use, instead of open API's LLM is instead of anthropic instead of others. So the Learn ln, just to be clear, there's actually a website you can go to, to kind of learn anything, and that is powered by learn LM. But I really think of that more as a demonstration project. The learn LM itself is basically offering API's for developers to build on it. And the idea is that it's safer, and less prone to hallucinations that are inappropriate. It also has like pedagogic knowledge and so on. Really interesting about this is the contrast between these inaccurate search results with Gemini, and then their positioning as you can trust, learn LM, I think that's going to be challenging. And remember, just a few months ago, they had the Founding Fathers kind of recast as African American leaders, in part because of some of their engineer prompting. So I think this tension between what's safe, what's accurate, how much do we want to constrain? versus how much do we want to have open? That is going to be a tension that basically every individual and tech company is going to have to wrestle with but also these large LLS? Yeah, one last thing on this is, we're actually seeing Google, Google has more of a faceless component to it. There's no singular character or figure that is representative of Google. And that's part of their ubiquity play, which is we're just going to be in everything, right? In fact, if you were looking at there was like so much to take in that it was clearly not intended for a consumer audience. It was intended to say, hey, here's all the products you already use, AI is going to be in. Yeah, whereas an open ai plus Microsoft, huge bets on Sam Altman, the figure that huge bets on Sal Khan, the figure, you know, we talked about this, in January, sales got a huge burden on his shoulders to pull the kind of promise of AI. And he and Sam are both very vulnerable to attack from the press, when AI does things that don't live up to their future. And often, you know, Sam and Sal are similar in that they're talking about over the next five years or the next 10 years, and people will then protect them on what the AI is or isn't doing today. So very, very interesting. The contrast. Yeah,

Alexander Sarlin:

that's a really good point. But at the same time, Google is also very sensitive to criticism, you know, I mean, you just mentioned that they went out of their way to do this trained AI to basically try to combat bias proactively so that if you asked for, you know, a doctor, or a teacher or a founding father, it would give you a person of color. And then people had backlash against that. And as a result, Google just shut down Geminis ability to make people at all until they figure it out. Meanwhile, this pizza and rocks scandal of the week, they are responding to that. I mean, they're so even though Google doesn't have a single figurehead and a person to attack. It is I think, acutely aware of the bad press that comes with AI of the nervousness that people have. Another survey about AI came out this week saying that, you know, a quarter of teachers feel like AI is going to do more harm than good, which is, you know, pretty decent number. I think Google is very aware, even though there's no one person to attack, you know, to persona one face to smack like, you know, Sam Altman, they are very responsive to the negative press. And you know, in terms of your point about integration, one of the things that is core to what LLM to learn LLM is it's integrated into Android, which is extremely important because Android, the Google App Store, and like the Apple App Store are home to hundreds at this point of AI tutors and various AI education products. And if learn LLM is going to be the go to API that's incorporated into Android. They may be making a bet that the app their app ecosystem can level up based on the better, you know the better pedagogy of their LLM. It's also integrated into YouTube, which is really noteworthy because YouTube has made some very public announcements in the last you know six months about about how it's really thinking about the education market much more seriously embedding quizzes, auto creating quizzes and creating courses. You know, YouTube has a course feature now where you can create a full course on YouTube. And it's actually designed as a course. So you can sort of see how that strategy like exactly what you're saying is is enrolling, you build the underlying infrastructure, tried to make a pedagogic LLM, and then fold it into all of your core products, which happened to be the most use things in the world, right, Android and YouTube are like the two most popular of their type in the world, as is Gmail, as is Google Drive. There's a lot going on here. Let's talk quickly about the sort of, I would say almost like the dark horse, the sort of independent party in this in this race, which is hugging face this week put out an announcement that they're sharing $10 million worth of compute power, to help people build AI apps to quote, you know, beat the big AI companies. We haven't talked about open source in a while, we've been focusing on all these big tech companies. But you and I put out an Article Six months ago about how people should be giving away compute, because that's really the blocking the blocker to innovation for a lot of smaller companies is the cost of compute and hugging face seems to have hurt us, don't you think that? I

Ben Kornell:

mean, they must be subscribers to the newsletter? There's no doubt. I mean, look, this is actually the dilemma in the AI race is that fax following is a great, great strategy. Yeah. And open models, whether they're fast following or taking the lead, and they have so much speed. And so then it comes down to who has the infrastructure advantage. And what you can see is that Google is making it basically free to use AI and Microsoft open AI are making it basically free to use AI. And their advantage is that they have the compute. And in some cases, they have the devices you just mentioned Android and and we were just talking about Chromebooks, that there's a way in which like, is AI really, actually the loss leader that sells the cloud services at that sells the hardware. Yeah. And so you know, for open models to thrive, what they really have to do is limit barriers to entry. And I think cotton face is doing a great job of basically shining the light on the quality of these open models, and then encouraging people to build natively open source. And you know, we do have the grok interview, I think there is this idea that a wrapper layer around open source could be incredibly powerful and affordable option, especially for third party developers. So this 10 million, I think, is a drop in the bucket compared to the total cost of compute in the space. But I think, especially for edtech, where the customers are very sensitive to price, and the developers are very sensitive to costs of compute. The open source community is offering some pretty compelling value.

Alexander Sarlin:

And freemium models, right ad tech often relies on both on the b2c and b2b side on freemium models, but AI sort of limits the freemium, you know, ability to really give the product away. If you're paying for compute sites, I think that sort of offsets it to some extent as well. So we've talked a lot about AI. Some of this is directly educational, but let's talk about some of the things that are happening in AI. And education. Really, specifically, we saw, you know, one of the biggest names in the EdTech space, Luis von Ahn, co founder and head of Duolingo, create, basically, at a conference this week talk about how AI is going to displace humans in many, many different jobs. And he's he's pretty clear that he is a computer scientist. And he is very clear eyed about where this is all going. But he specifically went out of his way this was as an education conference, but to say the teachers are one of the jobs that are going to be harder to replace, because they're sort of moral pillars, they'd model things for students, they're some of the most human workers, so to speak, in our entire economy. That was an interesting take, you know, obviously, somebody like Luis von Ahn, who has been working in AI for decades, has to be very careful about not going you know, full Silicon Valley Tech speak on his core audience of learners and educators. What did you make of that? Do you agree that teachers I mean, all of us want teachers to not be replaceable? And I don't think they are. But it was interesting to hear somebody in the space say it's so specifically.

Ben Kornell:

Yeah, I think we were in danger of losing some nuance here, in that replacing teachers certainly is not the right mental mindset. But supporting, augmenting and scaling teachers could very well be the path. And let's just also pair this with the new story we've been covering for nearly two years now, which is there's a crisis in staffing. That could be teachers that can be paraprofessionals. There's just a human capital gap. And then if we go outside the US, and you start talking about the global education ecosystem, there's full countries that have only 10 or 20% of the certificate teachers that they would need to serve all kit. So the potential for AI is, it's like most disruptive tech, it's not going to be better than the best teacher. And I think Dan Meyer had a really great newsletter talking about the limitations of AI versus a great teacher or even an average teacher. But AI as a supplement can either amplify the average or great teachers so that they can serve more students and be more efficient and have a job that is more sustainable. But also, they can raise the floor on those who either are under a credential, underprepared, or just straight up, have there's vacancies. And I think that's going to be our we don't really do well with Nuance these days in our highly politicized education space. But I think that that's what every industry is actually wrestling with. It's not this or that. It's how do we combine those things together? In fact, Ethan's Molix book is all about cohabitating. With AI, not a replacement mechanism, very

Alexander Sarlin:

well said and very politic. And I think you're right there, you know, augmenting word of the three verbs you use. I like those,

Ben Kornell:

we'll have to rewind the tape.

Alexander Sarlin:

Okay, yeah. Well, they were good. I agree. I mean, you know, I always find it very useful when thinking about the future of AI to look at sort of the internet, which is, I think, the last big tech evolution that sort of is remotely close in impact. And, you know, is the internet better than a teacher? It's kind of a ridiculous question, isn't it? Right? I mean, yes, you can get any information you want on the internet, you can also access any person in the world on the internet, it clearly has a informational and scale advantage over any individual person of any kind, that said, would you want your student you know, your own kids to be put in front of the web to get their education instead of you know, like, go sit there with a Chromebook or, or an Apple laptop and say, Hey, go educate yourself? No, you probably wouldn't. There's a lot of gaps there, many, many, many, many, many, many gaps there. So, you know, I think there's definitely room for them to play together. And it's gonna be it's gonna be interesting, I think, on the student side that another big story that caught my eye this week, there's a TechCrunch article about how AI tutor apps, we just mentioned them in passing, but AI tutor apps are getting incredibly popular. One called question AI has over 6 million downloads. There are a number of really big, I sometimes call them like skin unsolved kind of apps where you can sort of take a picture, like Photomath, but they're sort of newfangled versions, you could take a picture or a circle something in a book, and it'll explain it to you, but maybe give you an answer as well. And it's a little hard. Some of the biggest ones are from China. So your question is from Julia Wong, I'm probably mispronouncing that, but one of the biggest edtech companies, and there's an interesting play in here that I think we may be, I feel like I haven't had my eye on, you know, we have our database of AI products is now over 425 products listed that are all new AI products, since this launched a year and a half ago. But I don't think I've kept my eye as closely on this app ecosystem, which is just rampant. And as teacher can AI replace teachers, probably a silly question, but it can definitely supplement, augment them in various ways. Can AI help students? Yes, definitely. But it can also be a pretty solid substitute for their actual work or thinking. And because this is an app store that's totally commercial, there's nothing really stopping the capitalist steamroller from creating things that just make homework easier and easier and easier. This is not a new story, obviously, an AI but it's interesting to read that you know that the Chinese apps are really sort of lapping everybody else in this market.

Ben Kornell:

I think this goes back to my point around the number of qualified teachers, and also what our education system is designed around in, especially in countries where it's more designed around rote learning, AI is particularly good at solving those types of problems. And there have been a bunch of chatter on LinkedIn where it's like, if AI can solve your problem, like teacher, you need to rethink the question. And I think that's a bit reductive. Because actually learning how to calculate an answer is, is an important skill for people to have. And then second, there's a lot of the lot of the tasks that AI does, could be like, write a poem for Independence Day, that is a really effective future forward kind of task. And, you know, maybe you read three other poems, but it would be really easy to load those other poems into a GPT and then auto generate a poem. And that's where we're headed. And I just think that that's going to be make it even tougher on educators and education systems to differentiate what's the 100% Human, what's 100% Ai, because most of it is really in the middle. It's

Alexander Sarlin:

a great point. And I like your point that it is reductive to say, if you're writing an assignment that can be completed by AI, you're probably not writing a very good assignment that's probably over stated. At the same time. I do on some level feel like, again, using the internet metaphor, if you were giving students take home homework 10 years ago, or I guess at this point, like 25 years ago that said, Pick a country and fill in this worksheet, and you need to know their population. And you need to know where, what continent and you need to know what its main export is, people don't give assignments like that anymore, because they know, all someone has to do is type in a search bar, and they get all that information or go to Wikipedia, they get all that information, they don't even have to think about it, there is an evolution that is going to have to happen. It's pretty, I think it's gonna be pretty deep in terms of what pedagogy looks like, in an age where there's this collective intelligence that is, in our pockets in our Chromebooks in our iPhones, you know, they just announced it's going to be, I think there's going to be some major changes in how teaching happens, and how teaching training happens, what they look like, who knows, it's got to change. I think any teacher who thinks that they can continue to use the same, the four set of same assignments, same quizzes, same take home papers that they've been using for the last five years, I should probably think a little harder, I don't think it's safe to do

Ben Kornell:

that anymore. Yeah, I think, again, it's like not all the way one, not all the way the other, but it could be the end of homework. There's a movement to end homework for kids, kindergarten through third grade. And, you know, we may end up just having many of the assignments, you've got to complete in person so that there's assurance that you're not using augmented tools. My view on that with that, that's also reductive. Because we want kids to be successful in the real world where they would have access to those tools. But I definitely had my share of exams where we weren't allowed to use a calculator, and I could have used it at home. So totally

Alexander Sarlin:

productive is the word of the day, which is a sad segue to our next topic, which is that guild education one of the big unicorn $4 billion valued companies in edtech, reduced its workforce this week. Do you see what I did there by a quarter?

Ben Kornell:

Great. You're really played with that. I love it.

Alexander Sarlin:

It's actually obviously a really sad story. We know lots of people at Guild guild is an incredible and tech success story. So it's when you see giants like that, that have really found various semblances of big product market fit have worked with huge, basically been a middleman to huge companies to huge higher education institutions and have done a lot of pivoting as well. When they're cutting that larger chunk of the workforce. It's yet another kind of Ed Tech, wintry sign. We also saw Restlet, which is arguably edtech arguably coding, but replique cut 20% of its staff this week, as well. What do you make of the layoffs?

Ben Kornell:

I mean, it's so tough out there. I think we all know about the K 12 funding cliff, the enrollment decline and the fiscal pressures there. And then we also know about university enrollment decline, and some of the challenges with the ROI on colleges and universities. And inflation has especially hit universities hard when you're looking at their cost to operate. But this was the first really big workforce, you know, very, very large hit in the workforce base. And I think it goes to show that relying on companies to pay for worker upskilling as a sole source of worker upskilling. That is probably not a long term, reliable strategy. While I think most employers want to mirror skill employees so that they can do the next set of tasks in their existing job. I think it is hard for employers in tough fiscal times like they're weighing, do I cut another headcount? Or do I cut this learning and development l&d program, they're gonna cut lnd spend before they cut people. And so that makes the kind of beta of these companies according to the economic highs and lows, very, very challenging. Yep. If you compare that to with the change in the investor sentiment, which was grow at all costs, capital is free grab market share to today, which is profitability rules. And so I think there's also something around guild really growing forward into what they thought growth potential was. And this retraction back is also acknowledging like, we've got to get profitable, there's not going to be a big funding round of, you know, a billion dollars coming down the pipe. And then the third thing in this particular case, you know, we have to acknowledge the, the incredible and harrowing story of Rachel Romer, the CEO, who came out publicly with their story, she had a stroke, and has young kids and is really just trying to get her life back to full operational status. And so she has a successor that she's named, and the woman who's leading Gil, I can't recall her name. But that's a tough time when you have to make big moves like this. And so, you know, I think there was probably a degree to which it was like, let's cut once, let's really go deep. You know, Rachel's not coming back as CEO in the near term, like this is your show now. And and so, you know, my hope is that this is going to put guild on a good long term track in that space. By the way, I'm still bullish on like, specialized credentialing, things like nursing credentialing, or things like, you know, particular cybersecurity credentialing. It's the generalist ones where people are getting a bachelor's degree, or they're getting a master's degree in a general topic that I think are the hardest to make the case for the ROI for the employers. Yeah, and

Alexander Sarlin:

guilt had already changed. I think a good portion of its model from being that sort of go between, between traditional educational institutions and constitutional may be called alternative education institutions and companies moving more towards a sort of upskilling within companies. So you know, they famously work with, you know, Walmart and Chipotle and Disney, places like that. And I feel like the announcement I think it was last year was really about how they're going to do something pretty different and really making more like be the fuel behind sort of Walmart University and all the credentialing there, and all the the infrastructure to make that kind of thing work. So that, you know, inside Walmart, or any of these big companies, you can sort of find your path and move from frontline to manager and that kind of thing. That was my understanding of it. So it's possible that this is a belated layoff to sort of acknowledge that shift again, definitely, you know, be Joe Shaw, who is the new CEO, I'm sure is very closely connected to this cut. Nobody wants to cut workers and they're very smart people at Guild. I mean, I think you should anybody who's listening to this, who is hiring guild has been a place that has attracted very smart people for a long time. And just last for me. I totally agree with your point about l&d. It's really been so confusing, trying to follow how companies think about l&d. One of my roles at Coursera was working with Coursera for business, and going with salespeople to l&d departments of some pretty big companies to basically try to explain what Coursera is doing in it. And I was just constantly surprised at how much even in these big companies, very sophisticated financial companies, all sorts of companies, l&d still seem to sort of need to justify its existence still seem to be nervous about the chopping block or trying to find a sort of core strategy to make themselves look relevant, which is just like crazy if you think about it, but at the same time, it's not that crazy, because it is not core to the bottom line, as much as you can find connections, they definitely there. But it's not obviously core to the bottom line that continually upskilling and training your employees is going to be financially positive. We all hope it is you know, and there are definitely some stats that it is, but it's not obvious. And I think that really matters. Because when things hit the skids to your point when you know, if it's getting rid of people or getting rid of a guild contract, it's the guild contract is going to come first.

Ben Kornell:

You know, it's such a contrast to Europe, where the labor laws are such that it's really hard to lay employees off. Right, so we see apprenticeship models, we see l&d learning models, really thriving in Europe, where the employees you've got today are the employees you have tomorrow. And you're going to have to build capability and capacity. And you know, it's awesome when times are hard in and this is a generalization. But in Europe, you also can't say, Hey, everybody, you're going to work 10 or 20 more hours per week. There's real limitations on the production of your workforce. And so the only way to kind of up level your output is to uplevel your team. And so I do think that there's a coherence in Europe around workforce, l&d and workforce as stakeholders in their economic and university systems. Whereas in the US, I think we've intentionally created a market based system, which means that we're just way more subjected to highs and lows.

Alexander Sarlin:

It's stakeholder capitalism, right? I mean, it's not it's not just market based. It's a specific version. It's like University of Chicago, Milton Friedman market based where capitalism

Ben Kornell:

that's Yeah, sure. Yeah. Love it or hate it. That's what we've got. And, you know, I think there's probably a lot of people who are not loving it right now. Speaking of market based and capitalistic tendencies in our space, one of our big Stories from the last two weeks that I think has really flown under the radar is Bain Capital, putting in a bid to buy Power School, Power School is one of the largest student information systems in the US, my estimate would be like 75% market share, but then I talked to somebody else. And they were like, it's only 20% market share, and the space is still fragmented. So I suppose it depends on which state you're in. But then making this move is really interesting, in part because it coupled with some of their other moves, which include their acquisition or their deep investment of media, which is the largest Capewell furnishing company. And then their recent leading of the round with Magic School, which is public because of SEC filing, I don't think that that's had a proper press release. But being ventures led the magic school, Bain Capital, lead media, and I think it might even be a third group like Bain Capital, capital P. It's kind of like when you have Google and there's Google Ventures, Google capital and capital G, I don't really know the difference. There's a way in which there's a potential could coherent strategy here. That seems to be almost accidental. But the idea is that if you have a full stack of the school, from physical infrastructure, which is all of your furnishing and chairs, which can be which you can basically instrument to capture data, audio visual data for L lens, then you have an assessment system, which is the student information system, who's attending, what are their grades, all of that, and then you have an AI generative layer on top of that, that can be a really compelling trio as they kind of go boldly into K 12. So, you know, part of this is we just haven't had a lot of big m&a News, this spring, like we thought we might do GSV. And part of this is really interesting to see a new player in the space kind of coming in aggressively in the form of thing.

Alexander Sarlin:

Yeah, it's a really interesting point about the full stack. I mean, the last statistics I've seen in the K 12. Market, Canvas still has the plurality or you know, Instructure Canvas, then Google Classroom, then power school, but they're all not that far apart. They're all in like the 30s to 20. So the three of

Ben Kornell:

them well, and I would differentiate an LMS from a student information system there. So there's a way in which Canvas works well in partnership with our school, and PowerSchool and

Alexander Sarlin:

Schoology. And that's the LMS. You're right, though. PowerSchool is si s can work with other LMSs. Yeah, I shouldn't have explained it quite that way. You're absolutely right. Yeah, I mean, it's interesting that they're going after such a proven player and PowerSchool has been around a long time he's made acquisitions, it's, you know, it's definitely not a startup by any means of the imagination. So I think that's really interesting. That's where people are going, because we had predicted that some of the acquisitions were going to be of some of these sort of mid sized companies, you know, small to mid sized companies that were hitting funding issues and runway issues and could be scooped up and aqua hired and get proprietary tech. And that's not really what this is. I mean, this is a very proven player in the space. There was actually a bunch of interesting stuff coming out of the m&a this week. So let's just do a little rundown. I know we're almost at time here. But like, you mentioned, the teacher shortage, Zen educate. This week raised$37 million. This is a British company that focuses specifically on teacher and sort of connecting teachers to schools, not just in the in the UK, also, in the US, and I believe elsewhere. That's interesting. And it was a $37 million round, which is pretty high these days. That was really interesting to see. And that's very much in the teacher shortage world that we're all in. We saw practica, which is a AI avatar based tutoring company that makes these sort of tries to be very, very realistic avatars that become your tutors, and you pick you up a character and you talk to it, and it talks back to you. It's one of these, there's not that many of them that do full avatars. But one of these types of products, they raised $35,000,000.35 and a half million dollars and say the report is that they have you know, over a million monthly active users. They're all over the world. And they're making about $20 million in the last 12 months. That's at least you know what they're saying in these articles. That's another vote of confidence for some of the AI tutoring space, even outside of the sort of app space. We saw. I'm still reeling from this headline. I don't exactly know what to make of it. But I left school, which is a Abu Dhabi. It's a UAE based ad tech company is doing an IPO for over $500 million for 20% of the stake. That is a lot of money and a big company that we have not I don't think ever covered on this platform I've ever covered on this podcast. It's a big platform. I'd love to look more into that. But we will look into that in future episodes. But you know, first IPO in coming out of the Abu Dhabi this year, we also saw mastery skills collective acquired by ETS which is really, really interesting. That's two very different but also semi related types of organizations, both nonprofits. And yeah, I'm still trying to make sense of that. But it's pretty exciting because ETS has such enormous reach and mastery skills is one of the leaders in the portrait of a graduate movement. Yeah, it

Ben Kornell:

is amazing to see the global edtech community continuing to grow. And it may be one of those situations where global edtech is the first rebound from the tech winter. And as a, from a sizing standpoint, global edtech is still relatively small. But the opportunities may be nearer term there. And in the US, because of many of the factors that we talked about in the pod today. It may be a longer slog to kind of turn the corner for ad tech. So very interesting to see some of this activity, m&a activity and investment activity happening outside the US. Yeah, well,

Alexander Sarlin:

so much going on. I mean, the fact that you get to name the Bane PowerSchool thing, sort of as an app, like you said, it's gone under the radar, that could have been our headline this week. That's a big deal in edtech. And it was sort of like the eighth story. It just shows how much is happening. As we go into this, you know, hopefully quiet summer,

Ben Kornell:

then I don't think it's going to be quiet. I think this is going to be a back to school season to remember for good and for that. And you know, just so you all know, too. I'm on my local school board. Our budget meetings are happening this week and next week, and we're deciding what do we keep and what do we cut. So there's some potential bad news coming for some companies, as districts make tough decisions as colleges and universities make tough decisions as parents and families and individual learners and corporate so it's going to be telling three months here as we head back into, boy,

Alexander Sarlin:

a little ominous there. Well, you'll hear about all of it as it happens here on EdTech Insiders Week in Edtech. Effects are bigger, everybody, and we will see you at the Oakland B's show or an ISTE or at one of the upcoming happy hours. Bye. Thanks for listening to this episode of EdTech Insiders. If you liked the podcast, remember to rate it and share it with others in the tech community. For those who want even more EdTech Insider, subscribe to the free Edtech Insiders newsletter on substack.

People on this episode