Week in Edtech 11/15/2023: OpenAI's Agents "GPTs", Edtech Trends, Acquisitions and Special Guest: Johann Zimmern, Zoom One for Education

So check that one out. And for our deep dive today we have a really special guest, Johan Zimmern, who is the education strategy lead at zoom, zoom, the teleconferencing system that we all use these days, they just released a really interesting suite of tools for education called Zoom one for education, we're learning all about it, and what Zoom is doing in the

education and AI space. And you know, generally we're just coming off of our conference, we just put all the videos on YouTube, we're putting together summaries of every panel and some really neat video highlight reel. So keep your eyes open for some of those things. That was a really fun conference. And already things are continuing to change. Just amazing. Yeah. And

leaders. You have school district leaders, we had the Assistant Superintendent from Gwinnett, who started an AI school and AI High School four years ago, and here they are, they're like, Wow, we never imagined this would happen so fast. And then on top of it, we had educators and students straight from the classroom, really just talking about the day to day life and how K-12 Higher Ed workforce is all getting disrupted. Jennifer, Carolyn was there kind of kick

the day off. So good to have investors in the room as well. And I do feel like we're getting this kind of come together moment. And I don't know, folks who are around in the 90s. Remember, no child left behind. And it's got a lot of like negative hangover effects. But one thing that really did define that era was, you know, late 90s, early 2000s was, education was the bipartisan like issue where everyone could kind of get together and agree on things and agree that it needed work and

support. And then I think it fragmented around school choice and charters, and then further has fragmented around public versus private, but it felt like, whoa, this AI in education. It's this coming together moment where we all see the potential, we all see the potential downsides. And we know over the next decade, this could be a redefining era. And so it

was just really exciting. And Alex, I will just say so many people reference the online event, too many of them had been to the conference, our AI and edu and then showed up at this. And so it was just great to have our community together and in person.

would do that. I'm not sure it fully has, but hopefully this is you know, something that makes room for change. The last thing we should talk about before we jump in is some pretty fun ad tech news, which is that our Friend of the pod, Jason Palmer, who is the Managing Director and Founder of new markets, Venture Partners, just announced he is running for President of the United States. That was quite unexpected. And he has running on a sort of talent D platform, but it's really not about

education. Ben, what did you think when you saw that coming through your various feeds? Well,

be? To really think about transforming education, education impact, by the way, those of you who don't know, Jason, their fund is really well known for some of the best returns consistently bull market bear market, and they do everything from, you know, early childhood, all the way to workforce and higher ed, and both in classroom and out of classroom. I did also think about like, Okay, what is the issue that really Jason is trying to bring attention to?

And I think it's really around education leading to opportunity. Yes. And I think that what we've had, is this education battle over what's getting taught in the classroom. Is it right wing? Is it left wing? Are we like brainwashing people, and I like the idea that Jason's campaign, whether he's able to take it all the way to the White House, probably low

probability there. But if he's able to get this issue, reframed as education is a fundamental foundation for creating opportunity for all and delivering on the American dream like that, I think could be really, really powerful. I don't know when you saw it. What were your first thoughts? No, I

And I highly recommend the new markets Venture Partners website, because they do a really nice job of sort of showing all the connections between the different levels of education, you know, why they invest across the entire educational spectrum and how it should all be sort of considered holistically, I really liked the

way they think about that. And we had an interview with him about that a while back on the podcast, but I sort of saw it a little bit like as an Andrew Yang type campaign, it would be amazing if he got anywhere near the publicity of Andrew Yang, it's probably doubtful, but like, you know, Andrew Yang, sort of came out of this investment business, you know, very modern tech world and brought ideas to the forefront that nobody was talking about, like universal basic income and

the threat of automation. And you know, some of these big ideas that people in Silicon Valley talk about, and because they resist could talk about, but people in Peoria don't talk about. And I thought it was a really powerful, and he got a lot of smart people, especially college kids, frankly, you know, across the country to really focus on some different things.

I think that would be an amazing outcome of this, like you're saying if he brought attention to the Theses that new markets have been following about, you know, education is really about opening opportunities, it should be equity, and fair, but it also should be integrated and really designed for getting students a successful life. And it should not be caught in these political battles, not be caught in this sort of the cyclical stuff that we always try to break out of.

It would be incredible. And, you know, maybe, you know, some people were positing, maybe it's an attempt for him to make a play of trying to be the next Secretary of Education or some role in a future administration that would give him at least some, maybe Department of Labor well, because you know, he's in

platform. So it can be a very effective strategy of raising the profile, but a certain set of interests or an issue area. And what you end up seeing almost inevitably, as candidates drop out as the campaign goes on, and then they join the lead campaign. So we'll be following that story. That was a fun one. Let's get into the AI. Let's start with this just ever changing world of AI. I mean, I tell you, Alex, between our podcasts between our weekend

antics so much happens. So maybe you can give a couple of headlines that really are standing out to you in terms of AI and relevance to tech. Yeah,

So that instead of going to ask, you know, Claude or Chad GBT, or being your question, you might say, well, I want a specific outcome. So I'm gonna go to this specific agent. And that agent will be trained to do that, and also have access to special information that's been uploaded into it, it'll have access to or trained, it'll have access to

certain tools. And open AI just launched basically, this idea of custom AI assistants that people can build, they're basically building a GPT, marketplace, or agent marketplace where individuals can create agents that do specific AI enabled tasks. And some of the examples they have, that have already been done that have already been created by open AI and themselves are things like math, mentor, creative writing, coach, all sorts of things that have

educational use cases. And I actually had some fun, I went on this yesterday, I went into GB t plus, I played around, and I made a couple of GP tees. And I gotta tell you, it is so easy to make them. I mean, for one thing, it's the same process as regular GPT. You could talk in plain language, you say, hey, I want an agent that does this. And they go, Okay, so I'm thinking this, and it'd be called that and what do you

think? You go? Yeah, yeah. But it could be that and this and that you negotiate whether you train it, you tell it what it's doing wrong, you can upload whole books into it, you say, Hey, I wanted to know this. And this and this, I was uploading the hero's journey into it and saying write stories that incorporate the hero's journey, you can have it use Dali to generate images in line, it is

pretty cool. And this is, I encourage everybody listening here, anybody who is already on chat to be t plus two, go in and make your own GPT. It's a lot easier than you think. I mean, it took you know, less than an hour to make one. And then you can publish it, you can send it to people, and it is literally a customized agent. So this is pretty fascinating. And we'll

have to see where it goes. But I would be surprised if this type of work isn't something that we increasingly see as sort of part of the AI story, especially for education. What did you think of this agent stuff?

the AI universe in chat TBT. And by the way, all their other moves have been accretive to this, they have Solly into check, some instructions brought in whisper. And so basically, there is this unified place that will essentially be the kind of center of the universe. And then the second on the GPT. Front is the typical evolution is that you have a Cambrian explosion of ideas, new things. And then eventually there's consolidation of apps have power and so on.

And if you look at like a store, like Google marketplace, or Apple marketplace, the kind of top 500 apps account for like 95% of the revenue for that site. What's challenging to me with this new technological moment, is that suppose somebody builds a really great GPT. The barrier to entry to create a mimicked GPT is very, very low. And so I agree that maybe consumer wise from a b2c standpoint, consumer might not want to take the time to, you know, build their own, and there's no necessarily

incremental cost. But I do think that if you are successful, you invite so much copying and parroting that how they structure their site for users to make meaning of it and not get overwhelmed, I think is really important. And it does. Also, you know, this is their second attempt, they had plugins, and that was a debacle. Because the plugins didn't really work in the same effortless way. They were complicated. They weren't

intuitive to use. Then the last piece is, I think we look at open AI and check GPT from an American lens around Okay, which companies which marketplace how does this play in ed tech, by the way, Ethan Moloch was one of their inside beta testers on GTs, and it's released a new GPT so love that they've got academics who are playing around with this, but I get really excited about the developing world where literally to launch your own app was just impossible

for most communities. You know, and especially if they're able to bring the pricing down of, you know, in global markets have chat GBT access. Imagine kids teens, like people being able to build locally sourced AI products in a GBT way that gets really, really exciting. And you think about mobile payments in that universe, you actually have a lot of African based mobile payment companies that have turned out to be huge companies that were all enabled by these

kind of app store concepts. So I'm very, very excited for leaps forward, especially for kids and students and young adult learners for this room opportunity. I agree. And I will just say this is one more thing, though, that is going to make the American K 12 educator just simply overwhelmed. Like the sheer number of choices, options will, you know, we're going to be in a period I think of this expansiveness. And then there's going to be a period where we

start narrowing. And I just think we just got another like year or two extension of expansiveness, the kind of cost to bring a tool to market now is approaching zero, like you said, an hour, it's an hour of your time now,

that's gonna work? It's hard to imagine anybody making. I mean, these are famous last words, I'm not even gonna say that I have no idea how the monetization is going to work. But you make great points. I think it is really interesting for developing world, I think it's ridiculously interesting for the ability of teenagers and young people to basically make very powerful, dedicated applications. The big mystery

underneath it all. And we've talked about this a lot here is that because these are all built on open AI, there's this sort of fundamental assumption that even if you have the most amazing educational GPT, you can think of on open AI, it's still on open AI, which means that technically 13 year olds can use it for that, or lar can use it without parental permission. Technically, you know, the data is not secure, and anything you put into it can be used for

training. Like all of those things that come with a big general model, trickled down into these, at least for now. I can imagine a future where they would allow somebody to turn off training data or do some interesting things to secure it for a particular use case, like education or health or law, but certainly not anytime soon. So it's gonna be interesting.

videos and stories. But this is sort of a next level of user generated content, because it's basically like you say, an app store you can create, it's

integrated companies? Because so for example, the companies that started with the grading in edtech. Ai grading, they have a choice? Do we go into really deep assessment with peer and formative and summative and data dashboard? Or do they go horizontally and say, Now you can do assessment, but you can also do lesson planning to do

like classroom management. So the kind of normal rules of product development, which were mainly constrained by time, yeah, strained by capital, constrained by like, you know, things like investor focus or market focus, these constraints are all off.

It's like early web, it's like, oh, I mean, one of the ones I made last night, just because it's an idea I've always liked. This is moral foundations theory from NYU psychologist that basically says, you know, the left and right, political factions have different ways of sort of looking at morality, they care about different sort

of moral foundations. And I'm like, Okay, I'm going to upload all the papers from this person, and then say, if I give you an issue, you tell me how the left and the right would see it, and how you would sort of explain it to one another, given their moral foundations. It did it in like 15 minutes. And I'm like, Okay, this is just like, it's an idea. That's totally, you know, a framework I've always found interesting, but it's just that you can do it so quickly. And you could just do that on

anything you find. You could do something that writes comic books with a certain style, anything. When I

GPT gonna launch because they

know, in writing the payout? I think the article really is positing some of these things that were just saying about open AI, which is not really open, it's closed AI, is really the name of the company should be closed down? How do we find ways to make sure that AI has ethical constructs built into it and guardrails and so on? How do we use it to drive research and public good in ways that might not be in the proprietary interests of these companies.

And what she really points out is just the magnitude of investing. It's gone from 100 times magnitude in like universities and private sector to millions upon millions of magnitudes. I mean, it's a flop of magnitude. We've been learning a lot about teraflops and flops. And I think it used to be that, wow, it's this cush job with like a cafeteria where you can eat all this food and

smart people around you. And now it's literally like to run the data centers one needs to do to do these experiences and experiments, only a select number of companies worldwide, not even full governments have the resources to do this kind of work. And that is like a really big pivot point for our world that the most powerful technology and essentially the kind of convener of what is going to be future ready and what is not, are in these very,

very small tech companies. And she basically posits that it's time to double down our investment in higher ed. I will just say in healthcare 10% of all, you know, money in health care goes to r&d. And that's partly because the payout on r&d with a breakthrough drug or

whatever it may be. There's like this sense of universities can develop new technologies that can go through a commercialization process that comes back to the universities, we have just not in education spaces than able to demonstrate that opportunity. And in AI, in particular, this race for proprietary means that universities are getting cut

out. So I think if we're really serious about what Biden is saying, with his executive order, there needs to be money to back it up with leading institutions getting funding, and then those institutions need to make sure also, it's not just the elite that are working in their labs, and so on, and so forth. So many

high demand. And basically, the industry and the private sector is the only and not only the private sector, the fangs, right, the few companies that have huge, huge, huge pockets are the only people who really have the ability to bring together the compute power and

the people. And therefore, you know, she's nervous, I think rightfully that there's not a counterbalance within the universities, because the equivalent of like, basic research, you know, we have these universities with huge nano labs, and amazing, you know, facilities, it's starting to get so expensive, so quickly to even begin to do any of this that that university is already getting left behind she sort of raising the flag on that. Because, yeah, and it's a really

fascinating point. And I think it leads to a lot of the things you're saying, it also reminds me of, you know, what we were just talking about, about where AI is going, which is that because it's so expensive to do the underlying training and model, get all the data to train those foundational models, these dual use foundational models to use the government's strange phrasing is impossible for almost anybody but the biggest

companies. So what it sort of leaves for anybody else is like, yes, maybe the US could put 100 million 200 million 500 million or billion $2 billion back into the sort of concept of basic research at the university level. So that there may be could be an open AI or a model, that is the US model that is used in the Defense Department that is not from a private company. That's, I guess, a

possibility. But what it also means is that, for the most part, most of what AI innovation is going to happen, will happen at the second layer, the next layer down from these foundational models, because nobody can afford these millions of dollars, at least right now to train

by the government, UAE. And I do think that China is doing open models thrive in the US or thrive abroad, there is the techno optimists would say a couple things. One, they'll say the cost of GPUs is going to go down dramatically. In the next 18 months, I heard, Elon Musk just launched grok, which is Chatbot. And they'd spent something like 100 million just over a six month period to get

it going. But in six months, they built like fully functioning AI chat bot with a funny and regular mode, all trained on Twitter data that said like he was predicting like in the next 18 months, the price is going to go down precipitously because of increased capacity in the

market. And I think the second thing that techno optimists will say is like, you know, there's ample open models for people to do the testing, and do the research where you can test a concept, even if you don't have the scale or reach, you can at least do the research to prove out XYZ, you know, case, and if you look at, you know, some of the research coming out of Stanford around NPCs, and their ability to grow and learn in like a simulated environment that shows like, there can be

some pretty interesting breakthroughs. Yeah, well, we should probably dive into K 12 and higher

little more centralized. I mean, obviously, Alibaba and Baidu are the equivalent of Google and you know, Facebook over there, but At the same time, there's a really interesting thing happening where the biggest companies are getting this huge advantage. If the cost comes down, that might change. But, man, it's a complicated world. And the the other thing is that there's a really interesting article about extractive AI that I just wanted to very quickly

cover. I know we're talking a lot about AI here, but extractive AI in the 74 million are really cool article, basically, long story short, is that extractive AI is AI, that instead of completely generating completely new text, it basically returns information through quotations that are sourced that are accurate, that

have citations behind them. And that, you know, in terms of education, that is a very powerful model that instead of doing something totally new, which you can't really verify, or you know, sometimes makes up sources, or hallucinates, at least for now, it's actually really basically a super powered search engine, where it can put together ideas in a very intelligent way. But their ideas that are coming sort of as snippets and quotes, I'm curious what you thought about that real

fast? And then I promise we will move to K 12.

structured data. And so this idea of data extraction is linked to that, which is basically like, how do we take these mounds and mounds of unstructured pieces of thought work from students or video capture of the classroom or audio snippets? And how do we extract the insights, the kind of learnings, the lessons the content out of that to then put it into our data analysis machines? To kind of understand what students are learning where

they're at? Are they engaged? I think one thing that I'm seeing a huge uptick in is sentiment analysis and engagement analysis AI. So not only can we look at what's the content and how our learners improving on certain topic areas, or standards, but also like, are they at risk of dropping out? Are they feeling down and depressed? AI can capture all of that. And so it's not really generating, it's still generative AI as a technology, but it's not generating new content, per se.

It's more analyzing what are these massive amounts of content and pulling out the parts that are most relevant. And if I had to bet on areas of AI, I would say, that's one, and then connect it to assessment. And then these like teacher, educator, Professor tools that scale the human beings, those are all no brainer bats, because the interest of the human and the interests of the tech and the path of the technology are fully aligned with better

outcomes. Whereas I still think the learner copilot is the hardest one because the learner has near term incentive to get the answer, right. Whereas the instructor has the incentive for that learner to learn things. And so you have this like, you know, the user might be actually working against the outcome that the institution or educator wants. But really great article, encourage everyone to read it. And it dovetails nicely into K

12. Where I do think in our kind of K 12 headlines, we've got several, you know, one would just be man, this teacher shortage is continuing, and it's continuing to get worse. But the three big ones are homeschooling is rise from fringe to fastest growing form of education. fascinating article about essentially state by state growth in homeschool. It's in the Washington Post. And they were looking at school districts

and the change. And one thing that you have to understand is enrollment drives revenue for schools and school districts. And there is a way in which some states, if you've got a fixed pie, and somebody is doing homeschool, and it's not subsidized. It actually drives up revenue per student in that state. And then other states that have homeschool subsidize, it drives revenue away from public education. So it's actually really fascinating to

see it broken down this way. But really incredible momentum there. The second headline is Instructure, acquiring parchment for 800 million. Parchment has 15,000 customers who've had 165 million credentials created over the last two decades.

Instructure basically is looking to add quality and credentialing they bought, you know a company that we know called Learn platform around quality outcomes, and then they buy this credentialing platform parchment, and they are really trying and I do feel like this is a stronger play in higher ed, but it is across K 12. to I think Instructure is really becoming you know, one of the it's going to be like the Pearson or you know, Scholastic

of the feature. And then the third one is more smoke and fire from buy juice. It's confirmed now that they're selling epic. And I think the story in general is around, you know, the colossal flame out of buy juice and Indian ad tech darling. But the takeout kind of share with our ad tech community here is this has huge implications for our our venture funds, many of which are very exposed to buy

juice. And so as buy juice kind of tanks, and sells off assets and tries to stay alive, the kind of the vision where they were going to IPO at 32 billion that is getting further and further away. And it creates this like really trickle down negative impact on valuations and exit outcomes for ad tech companies. And so if you think this is just an Indian ad tech, or you think it's just k 12.

It's not this is colossal, colossal news out of buy juice, and the fact that they're selling some of their best parts for you know, huge discounts, ostensibly, I think is really, really concerning for our space. Alex, I went through a lot there because, you know, Kate wells, my passion beats or anything there that spike for you.

Florida 72%. But I think one of the big headlines there is that just home, the rise in homeschooling versus the rise in

private school. And you know, we know that public schooling went down, you know, by a million students during the pandemic, and we had sort of some ideas of where they were going, but this is sort of the real, the real story is that basically, you know, the pandemic made a huge rise in homeschooling, it was not that they went to private schools or and I think that's worth just thinking about in terms of what, you know, did the pandemic normalize being at

home? Did it normalize the connections between parents and students? Did it normalize ed tech, as a, you know, a way of learning as opposed to regular school or differentiated personalized learning? I think there's a lot to unravel here as we go forward.

more acceptable option. I think, you know, there was talk about homeschool 2.0 In the years before the pandemic were sort of going from its original conception as sort of religious families or people who were who had, you know, very specific reasons to homeschool to being something that was more widespread across races, for example, and sort of more accepted, but I think this turbocharge that I think the just the idea of homeschooling as an option for almost any family is, is now on the table,

whether it's totally normalized, and will be for a long time. I mean, who can say, but I do think it's interesting to see that I mean, schools have really not done a great job in this year of school board fights and school shootings and teacher shortages and mental health crises like any of these could turn around. But my feeling is that schools have just don't have a great PR, right? I mean, they're just there's a lot of

problems. And it doesn't feel like a place that's bringing people a ton of joy and excitement in many cases. And it doesn't feel like that much of a loss to a lot of communities to pull their kids out and do a micro school or homeschool or some pod schooling or you know, any of the variations of homeschooling that we've seen. Right. So I think it'll be around for a little while. That'd be my guess. And then the parchment. I mean, yeah, you're right. It's not truly a K 12

story, but $800 million. So that to me here. I mean, that's not a small amount of money for an edtech acquisition. That's a pretty significant amount. Instructure? Yes, they have learned platform. Yes, obviously, they have Canvas, which is their big flagship product and sort of contains everything we just talked to share in the CPO of Instructure.

And he talked about, you know, some of the plans for Canvas and how it's going to be sort of integrated with everything and they're trying to make it the place to go. But what's interesting to me about parchment 800 million, and this quote from the press release about increasingly learning happens through non traditional channels and Instructure believes this is a key strategic area of growth. So tying into what we just talked about about

homeschooling. I think part of the purpose of parchment as a platform is that if you can credential and reward people for learning wherever it happens, that begins to really continue to pull the teeth out of school and higher At traditional higher ed as the sort of monopoly of what education looks like, you can definitely imagine if, you know, many, many people who go to school in either k 12, or higher ed are doing it in Canvas in this country, it's the number

one LMS. If you can go to Canvas the same platform and learn from an out school teacher, or you can learn from a company or learn from an externship and still get a meaningful credential from that and still put that on your portfolio or on your LinkedIn, it really changes I think some of the calculations even more about traditional education. So that's my take on

that. And then nothing more to say about bite us other than it seemed like an interesting little side note that Duolingo was potentially expressing interest in buying epic, I have no idea what to make of that. But that was in the press about it. So did you that jump out to you?

And I'm sure that they're feeling really, really confident around their performance quarter by quarter, but also wary of other people kind of coming into their zone, leveraging generative AI for language learning. And so what I love about this news is it just sounds like they're keeping their foot on the gas pedal, they're not getting complacent, you know, Duolingo, once you IPO, often, you lose the edge. And Duolingo seems to be like the total lingo

ed tech ecosystem. And so I just hope whoever has the most money is also like, incredibly strategically well positioned to bring epic to many, many more kids for learning and literacy, and also to combine it with other offerings. Yeah, let's do higher ed, we're kind of doing our around the world. So you know, we go k 12. Higher Ed workforce, what are you watching in higher ed? Alex?

from me at the same time. The opposite, you know, you could say it's by Jews, you also could make a pretty decent case that it is to you, as of this moment has bounced back a little bit, but it's been trading at under $1. For recently, after a series of sort of strange headlines, USC, which is two years original partner pulled the contract from them. And they're paying some payouts and $40 million payout.

But basically, they're ending the relationship and saying, you know, some of the things that to you, you know, investors and to your watchers have been nervous about for a long time, which is that this original to you model is starting to get a little old, and it's not necessarily doing the same thing as it used to do. And I think there's a people are getting pretty nervous about what's going to happen to you, especially with the degrees model. We're also seeing a little bit of a slowdown in

coding boot camps. Class central reported that to you was going to take over Pearson's OPM business, which is interesting and potentially a positive sign, maybe, but men, it is just pretty brutal. And as somebody who you know, I worked for trilogy education, they got purchased by to you. And so I was working at to you for a while. So I've worked in there, and I'm not, you know, I don't want to say anything bad about

it. But it is just really wacky to watch that company's, you know, fortunes be battered and battered and battered. You know, their stock was at over $90. Since the IPO in 2018. I think that was as high as 90 something. And now they're sitting there at, you know, one and $2 for weeks on end. So that is just a higher ed story. You could read it lots of ways as the sort of slow death of of OPM is as a business. You could read it as to you as a company just

having some troubles. You could read it as higher education institutions starting to want to do more of themselves or to have different ideas, but I don't know that one jumped out to me. It before we move on. Do you have any comments on that to you?

great. And then they bought trilogy, which was bootcamps 750 million units was doing that was going bonkers. But to you had to borrow a bunch of money to make that acquisition. So they bring on some debt. And then by the way, there's like, somewhere in that zone, there are these scandalous articles about them charging like 50% 80% margins on like MBA degrees at USC and some other universities where basically low income people are way way overpaying for what

they're getting. And you know, and essentially, it's very cynical around, you know, the real value is the credential, not the learning. And then they bought edX out of MIT, which was like the Coursera, competitor out of MIT. And they raised more debt on that, you know, we were in this high time where like, debt was cheap. And like, Ed Tech is like going through the roof. And as soon as they started heading down, you know what that does, when you're on a downward slope, it accelerates

your momentum down. And so they're making all these like debt payoffs, it creates real risk for the business. They're also just getting eaten from a lot of different sides. And so I'm wondering, like, do the edX people think, God, thank God, we sold it when we could? Or are they thinking, man, we had something really amazing and beautiful, and it's gotten totally screwed up now? Are they still hopeful for the turnout?

It's hard to imagine the kind of plane going down at this nosedive trajectory, really pulling it up and out and reaching the same heights that it was before? I don't know, how do you think that like? I know, you don't know that x people personally, but how do you think they would view this?

now non exclusive with us. And the parent company is obviously going through all this. I don't know, none of this is super insightful. But I would think that basically, anybody associated with you, they also bought gets smarter, you know, a few years ago, they've made some, I would consider them quite strategic acquisitions.

But it doesn't seem like they've actually made any internal pivots in the business to make those acquisitions the center of the business, and they've still been so reliant on their original model, and their original model has been really hurting. And it just seems like they made I think, acquisitions that I can't fault them for any of those acquisitions, I think they're very smart companies to

buy. And the vision of putting them all together under one roof does actually make sense, at least in theory, I think in practice, they are clearly have not been able to meet their earnings and have not been able to convince the world including their partners and the investment community that the combination of these companies is significantly more than the sum of the parts and that they're sort of have been moving quickly enough internally.

That's all I can think of, I don't know if edX feels like a pawn in that game or, or that they're, you know, feel like they you know, they had their run, but it's a crazy thing to watch a company that big, really struggle I remember when I was at Coursera and Coursera first went into the degrees business, which was you know, years ago now, probably 2015 2016. Two, you reached out and they were like, Okay, it's on, you know, you're coming after our OPM

world. And at the time, it felt like to that felt scary, it felt like to you was the big scary dog. They had been doing this for years, and really successfully, very, with tons of universities. It felt like oh, wow, you know, we're on their radar. Oh, and now, you know, it just seems like they have just made a number of what turned out to be missteps.

something. But the reality is it's tough to run these businesses, but also like what happens with do you what happens with any of our public market companies, because there's so few of them really has a trickle down effect on the valuations and investing activity in our space. And so, you know, this definitely not a cause for celebration, it just is gonna

prolong our ed tech winter. I think, unfortunately, let's go to workforce where, you know, we've had a little bit of a thaw, if you had to say, what's the one area that tech that still continues to go fairly strong, its workforce, I think one of the dynamics about that is, you know, retaining high quality talent continues to be a huge priority for most

companies. But then also, the rate of change in the marketplace means like, continuous upskilling improvement, and, you know, making a more nimble workforce is also essential. So the market dynamics have been good. What are you hearing about? And what are you focusing on on the workforce side? Yeah,

training. And given what you just said, which is totally true about the small number of public edtech companies and small number of sort of household name edtech companies, I think you're gonna see a lot more, you know, Duolingo, for blank pitches. But the idea of workforce as sort of a little bit of a safe haven in this time, still seems to be

going strong. Udemy just put out their workforce trends report, and one of the things they call out and maybe this is not super surprising is that the types of skills continue to that people are looking for continue to evolve very quickly. And they, and a lot of them are really about a specific, you know, tools, specific technical tools, so, chats, UBT, number one on their list of surging skills over a 4,000% increase, but of course, it didn't really exist a

year ago. So that's that, but you also see things like, you know, Google Cloud, DevOps engineering or Microsoft Azure, you know, cloud engineering, certifications, Dev SEC ops, that's, you know, a subset of coding words about like security operations, or Autodesk. And I always am sort of amazed at, you know, when you look at workforce training, yes, there's some sort of soft skills in there, or like, you know, some advertising strategy or, you know, comes on

that list. But so many of the things that people need to succeed in the workforce right now are like, incredibly specific, semi technical, or at least, at least semi technical tools. And yet, just does not find its way into any other form of education, which is just mind blowing. And I mean, there's another article this week about how universities can't accommodate all the people trying to be computer science

majors. And they're starting to Michigan is leading the way on this, but a lot of people are starting to basically create additional applications, you're already admitted to the university, then you have to apply and qualify for the major for being a computer science major. And you put these two headlines next to each other, and you're just like, What the heck is going on here? People are trying to get technical skills for these high to get great jobs. And everybody's

saying, Yep, you need them. And here they are. They're things like cloud development and Dev SEC ops, and there you go. And then they're going to college and trying to get in computer science, which doesn't teach those things anyway. And they're not even able to get in because if they don't have good enough grades, in freshman year, they're kicked out of the potential to even major in computer science. What a ridiculous situation, what a crazy mismatch between what people are trying to do, and

what is actually happening. So they're forced out of computer science, they become maybe an engineering major of another kind or a psychology major, and they get out and then they need to start from scratch and learn Azure, Google Cloud, and learn chatty, btw, like, I know, are people listening to this, I'm preaching to the choir here, and a lot of amazing companies have been working to, To remedy this, but it just drives me nuts, to see these sort of parallel paths

continue to roll out where you, you know, the traditional systems just struggle in these absurd ways. And the workforce just rolls on and asked for more and more training and more and more lifelong training and continuous development. It's just weird. But yeah, I mean, people who are looking to do things for workforce or for technical training or for authentic, you know, training for the skills that people really need. It's still a really hot space. I

different ways. I think the other element is like, what is the degree to which our university systems are actually well suited to be adaptive to the most cutting edge technology versus nimble boot, you know, market forces based Boot Camps and Online Programs. And, you know, my thinking is, this is part of the reason why the kind of ROI or the four year degree is being questioned, because you're going to get training for

a future of work. In an institution where many of the people training you have never worked in the real world, or when they did, it was so long ago that that knowledge is is relatively updated. On the workforce side, too. I would just also say, you know, I think companies, they're pretty solid. Now you don't see as many layoff rounds coming. And so that's

true. Many of these companies are looking around the room and saying, Okay, these are the workers, we've got, like, how do we get to where we need to, with these workers, and before they we're spending lots and lots of money on hiring and recruiting, but those headcounts aren't any longer approved. And so some subset of that hiring, recruiting spend, is now going towards training upskilling, you know, helping make their employee base more efficient.

You and I both got an email recently from a CTO who was saying, look at all of my entry level employees, I needed to do six months of hardcore training just to get them up to a level where I could count on them. And it wasn't just technical, it was like professionalism and workforce readiness, and all of

that. I think that, you know, from our podcasts today, you know, AI is certainly popping, but I do think that workforce is evolving into a next phase, which is less around every employee for themselves and much more around like, employers getting in on the game hardcore and saying, Okay, how do we get our workforce to meet the needs of the coming, you know, five to 10 years? Yeah,

move along. Another one I found interesting, partially because it's coming from our ventures, which has been sort of being very careful with its big fund is in a company called small, it's, it's a $4 million round just a couple of weeks ago, it's really about the talent supply shortage in the green energy transition, which I think is a really hot and very interesting space, and very aligned to sort of the goals of the next, you know, the younger generations

right now. So that's a Berlin based company Alaba green energy, but basically, you know, future proofed climate jobs. Super interesting to me, what stood out to you,

powerful area. And one interesting thing about this as well as for four year olds, you can invest in b2c motion, but you could also do a b2b motion now that many schools and districts offer you know, TK or you know, four year old so that one was one that caught my eye, and then classroom and India as a school infrastructure company.

And I think that that is really basically school infrastructures becoming a hot space because if you're a man imagining the vertical stack and the LLM being your infrastructure play. There's also some infrastructure plays around instrumenting the classroom so that you can basically power voice and speech

AI. And classroom is one that also is connecting online and offline modes of learning, which I think you know, that's another really interesting infrastructure play that, you know, it's not all online, and it's not all offline, and that you could actually have connections between both. And just given the scale in India, one thing that I've heard many times that there's literally not enough seats in the physical classroom to meet the full demand. And yet the online quality is also not meeting the

quality bar and outcomes. So I love this idea of like a bridge that basically connects the classroom and online and back and forth. So it was a small round for under $50,000, but definitely caught my eye and it's an area I've been looking at. For

education strategy. Tell us a little bit about zooms relationship to education. Absolutely.

education program. And we realized early on that educators and administrators needed tools to communicate more effectively. And while yes, Zoom is used in all industries, education is one of our largest industries still to this day. And that was exponentially increased. Obviously, during the pandemic, we started as a company in the meeting space and have since

broadened it out our scope. And what we call unified communication as a service is now the focus on this next stage, this post pandemic stage, where we combine everything from chat to telephony, to webinars, events, Contact Center, and most recently, the introduction of AI in our tool set to improve the communication stack even further.

us about what that is. And you know why there is this, you know, now we are sort of coming to the tail end of the pandemic, what were some of the big learnings during the pandemic presume, around the education use case that are being sort of baked into this zoom one for education product? Yeah,

trips. It's all these things that happen in many cases, even prior to the pandemic and are now being continued school board meetings, administrative gatherings, because a lot of employees are mandating a flexible work schedule, they want to work from home, whether they're at university or a school district, the individual institutions have to kind of build out an entire technology stack that not only includes zoom, but it has to be connected to displays, microphones,

speakers, cameras, what was an emergency remote education environment is now kind of settling into being a hybrid work and a hybrid learning environment going forward, probably a little bit more pronounced in higher education than K 12. But it continues to be a necessary tool to just facilitate work and learning in

general. And then with that comes the the introduction of Unified Communications, which in our case means not just the meeting portion, but we're trying to figure out how to build our product portfolio so institutions can purchase under a single umbrella and zoom one was launched probably a year and a half ago in the Marshall Space. And we took our time to really understand what does zoom one mean for education? And how do we want to make that available. So we have two

different options. One is called the school campus version that you can buy as few or as many as you need. And then we have enterprise wide coverage models for schools and universities that want to cover everyone at their institution. And it really combines chat, video, audio telephony, in some cases, webinars, which previously were very expensive, and you have to buy them for some of your individuals are now available to

everyone. We include translation transcriptions, so if you have a school board meeting, you have individuals in your district that don't speak English, necessarily, or don't speak English, well, they can turn on their own language on the subtitles in the captions in a live meeting. So not only do we drive it on the back end, but we actually provide you with AI translations, ad hoc,

And it's interesting hearing you talk about the hybrid world that we've sort of all inherited post pandemic, where, you know, education itself, has hybrid models, there are people still learning from home, there are people doing a few days a week, in the higher ed space, there's every different combination of people, you know, learning

online. But beyond the actual learning, you've mentioned, a whole bunch of interesting use cases about, you know, school board meetings, and parent teacher conferences, and all sorts of communication moments that happen within a learning environment that did not may not be a pedagogical moment or teacher student moment. But it still is incredibly important for getting everything together, teachers meeting with each other, for example, or department meetings in a

college. What I'd love to hear about is, you know, as you spent this year and a half thinking about how to adapt the Zoom one commercial to zoom one for education, what were some of the specific use cases in education? And now I'd love to hear some of the AI as well, I'm sure, obviously, AI is being used in

this translation. But how are you thinking about the education use case and using this incredible new technology we're all grappling with, to make zoom one for education even better and continue to accelerate with AI,

meetings. And very quickly, we realized, oh, we need to actually tie that to the learning management system. So we use LTI 1.3 integration, to ensure that not only your meeting is integrated with the learning management, but your whiteboard, your separate whiteboard is also capable of just being attached to all of your coursework. And that is unique in education, where we obviously look at different industry, manufacturing, retail health care, they all have different needs, but education

has unique needs. And that started even during the pandemic, teachers were saying, Hey, I don't want my students to see each other's videos, because it's distracting. And some students don't feel comfortable sharing it. So we created a focus mode feature that allowed the educator to see everyone but the students couldn't see each other. And that was extremely helpful. And there was a specific feature that was

requested from educators. We also added American Sign Language Interpretation windows that a student could pull out of the meeting window and saying, oh, I need my sign language interpreter separately, in a different window, so I can follow along with the meeting. So accessibility is incredibly important education. And we look at these features individually.

We have an advisory board, actually several advisory boards, both in K 12, and higher ed, and internationally in order to get that feedback, right, what is working for you as an administrator? Why are you purchasing a particular platform? What is important to you and how does it all tie together? And in many of our conversations, we find out that the IT administrators and procurement would like to have single source purchasing abilities. And they'd like to have tools that actually

integrate nicely. So we have team chat as part of zoom. It's a separate chat standalone application that is competing with Slack and discord and many other tools out there, but because it's integrated in the Zoom experience, people are adopting that at greater rates. Because of the integrated experience. We've integrated our emails You know a lot of things to get away from the toggle tax, I have to open up 17 windows to do my basic work

externally on websites. Imagine the student coming to university website, say I'm interested in engineering and they have a starting conversation with a chatbot and trying to figure out where the links are on the site, then they want to talk to a career advisor. And they said, connect me to an agent that can start as a chat conversation that can be elevated to a phone call. And then even further, it can be elevated to a video conference where the advisor can say, Oh, let me show you around.

Let me give you a tour of the school. So trying to tie the AI with the capabilities of the institution with a human connection. And one of our slogans is providing limitless human connections. It's really about that frictionless engagement and the satisfaction on the student side. We also added to our telephony platform contact center capabilities. So now you can have contact center capabilities for your IT support for your health services for

enrollment recruitment. It's amazing how the pandemic changed the expectations of individuals of how they want to connect with an institution, it used to be, you gotta get in line with 20 people ahead of you, I want to say my English one on one class, oh, by the time I get to the desk, sorry, classes gone. That's all moved online. And the communication with faculties and administrators has gone online, we recently added scheduler

capability. So it's similar to Calendly, where the professor can say I'm available Thursdays from nine to 12, pick your 15 minutes left. And it'll automatically create a calendar with reminders with a zoom link, it makes it so much easier, people are more productive, they don't have to travel around campus. And the virtual services side when we consider students services has really grown dramatically, including alumni communication. Now we need to do webinars, we need to do large

events for alumni. So we do want to do fundraising events. We added zoom events that allows you to do multi day concurrent sessions with registration with monetization if you wanted to charge for attendance. And it's really broadening our entire platform play. Some of those tools are inside of zoom one. Some of those tools are separate. But we have gone far beyond just meaning.

I was a presume, undergrad. But now, you know, I'd be sitting in the coffee shop on a phone, talking to my advisor, you know, and it makes so much sense in this always connected world. And with these huge campuses and with these hybrid models, it's really interesting to hear you talk about all the things that AI is doing inside zoom in all

the places it's going. But I'm sure some listeners may not realize, and I didn't know all of this as well, that Zoom is such a you know, when you talk about unified communication, that that includes whiteboards, you know, as a service that includes emails, that includes chat, you know, people Zoom is so considered so synonymous with teleconferencing, and, but it actually has a huge number of different communication options within it all of which can be

used in separate ways. I love the Zoom whiteboard, by the way, and the annotation feature, because I'm on zoom all day, as are many of us, I often surprised people by sort of right drawing on their screen while they're sharing and things because you can do that with

annotation. And I really think that Zoom has sort of enabled types of communication and types of pedagogy like, you know, collaborative whiteboards, or this kind of focus mode that you're saying where a teacher can be monitoring every student, there's no back of the classroom, there's no you know, people passing notes, but the students are all focused on on the teacher. That's a very different learning experience than we've ever really had in

the past. I'd love to hear you talk just a little bit about you know, as you're thinking about all these features, and the analytics, the transcriptions, there's so much in a package like this that has calendar services, it's just like it's a really comprehensive package that I'm sure is being sold to higher education and to schools and to districts. You know, how do you think about how zoom has

actually changed pedagogy? Like, what are some things that you've noted from talking to your advisory board or some of your clients about? What's changed about teaching itself in this zoom era?

you're using. But I think what's important is we're now enabling multiple learning modalities in a way that students are now raising their hand and saying, I don't want to be in class at 730 in the morning, in the lecture hall with 200 people, let me log in online, so I can watch it from my dormitory. And I recently talked to some kids at UC Berkeley, and they said, I love being on campus, I want to be with my friends and my cohorts, and want to collaborate

with other students. But I don't necessarily want to be in every single class every single day. And I want to have options. So let me watch the recording on the weekend, when I have time. Let me watch it live even from my dormitory, because I'm too lazy to walk across campus to get to. And in some ways, it's you know, we can understand why students want that flexibility. They pay a lot of money for tuition, they want to be served the way they want to be served.

And it's a little bit of a struggle for faculty to address that. We've had the term high flex that was termed here in San Francisco, by Brian Betty. And, you know, that is probably the ultimate idea of how do I serve everyone? How do I serve the students in class? And how do I engage the students that are live online? And how do I then also provide the same learning capabilities in the same learning outcomes to somebody who wasn't in class, and who's just watching the recording

afterwards? So by doing including things like transcriptions, right, we record the session, we provide transcriptions, we provide summaries. And with i Now that takes it to a whole nother level, where I can have a two hour session, and AI will summarize the class so to speak, or the administrative meeting, and provide that information to everyone. If I join a class late, I can use AI to say, Hey, what did I miss in the first 15

minutes of this call? And AI will help me summarize what was discussed prior to me joining. So there's now new ways to engage both administrative and the learning side. But I think on the learning side, there's things that we haven't yet even seen and imagine, because of these new capabilities.

app marketplace. So one thing that's come out in the last year, maybe a little longer than a year, is this concept of people being able to sort of create apps and plug into zoom conferences, and there have been a few edtech companies that have really taken advantage of this. The one that jumps to mind, of course, is Kahoot, which is one of the featured apps on the Zoom app marketplace. It's you know, pretty universally beloved education app for engagement and being able to sort of assess on

the fly. Tell us a little bit about your take on the app marketplace. from an education standpoint, is it something you encourage edtech companies to be leaning into and building for? Or is it you're working with the Zoom API's? Or is it only certain companies? That would be a good match? How do you think about that world?

And what I said earlier is like unified communication is all about not having to jump around and grind to different applications, I can make a phone call directly here from my desktop, if I need to talk my doctor, I'm not picking up the phone, I'm just dialing directly from here. If I want to text my children, I'm opening up the text feature in my Zoom client to send a text message to my

kids. And the apps are just a further extension of broadening the ecosystem of what we can offer and tying in educational content without the individual, the teacher the faculty having to jump around. So if I want to use Kahoot Yes, I could use code in a screen sharing mode, and I can open up a separate window. But wouldn't it be nice if it was just integrated, I can open them straight in the app. And whether that's us, or Prezi, or many, many other applications. It just makes it easier and we

encourage that development. And we also provide SDKs and API's for developers to build on top of the Zoom plot. Forman the great example is class Michael chasen who founded Blackboard, created an entire company based on the Zoom platform saying I can extend this even further, I can create even more features and make an entire product portfolio based on the Zoom platform. We

and enterprise models. In integrated comprehensive LTE eyes, transcripts, translations, chat, mail, calendar, phone, scheduling, access, of course, access on phones and tablets.

It's an exciting vision. And I'm sure that you know, some of our listeners who work in the administrative space are probably already have been thinking about zoom, one for education, but just the idea of realizing that Zoom is so deep in the education use case I think is really something that's important for us in edtech to know that a mega tech company, you know, sort of synonymous with teleconferencing globally, is really bending over backwards to make sure that education is

served the way it should be served. So I really appreciate you being with us here today. Johann Zimmern education strategy lead at zoom. Thanks for being here.