McGraw Hill, Inc. ($MH)

All right. Great. So we'll get started. Happy to have at the conference for the first time, McGraw-Hill. And with us, we have President and CEO, Philip Moyer. Philip, thanks for being here.

Thank you, David.

So Philip, you're about 3 months into the CEO seat. Maybe can you briefly cover your background and what attracted you to the education sector in McGraw-Hill specifically?

Sure. And thanks very much, David. I'm really excited to be here at the conference. Great conference. Really looking forward to meeting with all of you as we go forward. My own background, just real briefly, I spent roughly a little bit almost 25 years in big tech. So I was at Microsoft in the very early days before it was worth -- before it had about $1 billion of revenue. I was there for a number of years. I spent a lot of time at Amazon, AWS as the cloud was really coming into its own. And then most recently, I was at Google, I was Vice President of AI Engineering, the Applied AI Engineering teams. Also just prior to this, I was CEO of Vimeo, who was sold to Bending Spoons. And one of the things that attracted me about education, I kind of -- you can't kind of turn a corner today or speak to somebody without having artificial intelligence being talked about. And the reality is that today, the average child that we're educating is probably going to live -- could live to as much as 100 years old. And so I thought it was a very important time to be working to spending as much time on human intelligence as we are on artificial intelligence. Prior to my joining Microsoft, I had a small start-up of called Orion Systems. We built software for the individualized education programs and special ed. And I dump that one day, every child would have their own individualized education program. And we are literally at an inflection point right now where both technology and pedagogy are arriving between the confluence of data, between pedagogy, between content and between tools, we're now actually able to start giving every single child an individualized understanding of their proficiency.

So maybe we can unpack that a bit, right? So your background, you just noted comes from high-growth tech platforms. I guess from that lens, kind of what are your initial views of the education industry? Where do you see the big kind of unlock opportunities ?

So now more than ever, in every single stage of education, data and analytics is arriving in education. Now there was a really important article this past week in the New York Times around literacy rates that now only 37% of the students in the United States are at or above proficiency in literacy. It's a crisis at this point. And over the past few years, roughly about 10 years, 87% of states have moved backwards in literacy. That's a crisis. What we're seeing in higher education right now coming out of college without a career, with loads of debt, outcomes-based focus in every single element of education is arriving. Data analytics have shown up in every single industry, whether it was the financial industry, the retail industry, we have an outrageous amount of data and analytics. Data and analytics is arriving right now inside of education. So that's probably, first and foremost, 1 of the most important things that you should realize is that it's becoming much more focused on outcomes. The second thing is there's a mass proliferation of tools right now. In the average K-12 school district, there's roughly about 2,400 tools, learning tools. Teachers and students have tool fatigue at this point. And they're looking for simplification. The last thing I would say is that there is an expectation of personalization. We have it on a day-to-day basis in our own lives, and I think students are expecting a personalized experience. Those 3 things are very important influences right now. It's one of the exciting reasons why I joined McGraw Hill, is this opportunity to just rethink how we're approaching education.

Got it. Phil, you recently wrote a piece in Fortune addressing AI fears and comparing it to prior tech panics. Maybe can you walk through your framework and understanding the challenge of AI and LLMs as it relates to education.

So I've been through this far too many times at this point. And literally, this cycle is following the exact same pattern. So I was around at the birth of the software industry at Microsoft. I was there before there was a word and before there was a Microsoft office. And it follows this exact same pattern where Bill Gates, I think, famously said early on, when I was there, I said, "Boy, if my software didn't have bugs, I wouldn't need salespeople. And I think every generation, whether or not it was the original -- the software industry, the birth of the Internet, the birth of the cloud, when cloud came about, we said we're not going to need any more IT people. We have more IT people than ever. Today inside of AI, we said we're not going to have software developers. We're not going to have inside sales people. We're not going to have inside sales. We're not going to have customer support. Consulting is going away. Last week, I just started a services company. It's hard. The last mile of every technology is hard. Keeping guardrails around these things, keeping these things on task, making sure they're accurate, making sure they're not hallucinating. All of that is very hard. In education, it's extraordinarily hard. We have to go through cycles in some cases of years to get a piece of education approved by a K-12 district. We have literally dozens and dozens of teachers in a single district. -- that have to evaluate and see whether or not it can be put into the workflow and whether or not it has efficacy. We have to send our tools through third-party assessments to really make sure that they're meeting the standards that are required -- and then literally professor by professor in higher ed, we have to help to integrate it into their syllabus, so they can teach. And so if we bring new case studies out in the course of a particular semester, we've to actually help them integrate it into their content. And so this last mile is very human-oriented, and education. A human is the actual outcome is the thing that we're building, human intelligence. And when you get humans involved, it's very difficult. And in artificial intelligence, I think it's a wonderful tool. We are using a significant amount. We think it's going to help us accelerate content. We think it's going to help us be better at what we call precision education, and we also think it's going to help with personalization. But there is a need for the last mile to be able to integrate it into the human workflow and into the educator workflow.

I'm going to circle back to the last mile, but in the fortunate piece you made some other great points about just sort of the uniqueness of education, the different learning states that people come in, also the energy required, right, to power an models. Can we touch upon these points?

Yes. So I said inside of the article, the average AI model has about 1.7 terabytes that sit in it. When you train one of these models, it becomes completely static. You literally have a static model. It's like running a program. It's like running a database query. The thing is locked in memory. And you run these things, the training cycles that you have to build for these things are extraordinary. And yes, they absorb a lot of data, but they then lock into this 1.7 terabyte to 2 terabytes worth of content that's kind of locked in. The average human brain has that in about the grain -- what's in the grain of rice on the size of your mind, has roughly that much data that sits inside of it, the synapses. And then as you learn, every single night as you're sleeping, what you're doing is you're essentially storing all your memories of that day. Even during the day, your brain actually has its own unique shape. It has its own unique understanding of education. And so student-by-student, there is a different physical representation of everyone's knowledge in their mind because of Synapsis coil. And I joke that unlike an AI model that requires Three mile Island, we can train an 8-year-old on a couple of tomato soup and grilled cheese, and quite frankly, we do believe that for the next 100 years, we can make sure that humans are actually better than the average AI model.

Got it. Maybe going to your point about the last model maybe you can unpack the factors that would make it difficult to replicate the value that a company like McGraw Hill brings to the classroom. Maybe touch upon that a cumulative student data part of it.

So I really encourage everybody to understand what's happening right now in what we call the science of reading, and I'll spend a little bit more time on this later, but it's a foundational issue that we faced. The reason why literacy rates are collapsing is because of what and how we have taught. So the content that we've been teaching and the way that we've been teaching has turned out to be foundationally, fundamentally wrong. We've been using what's called a whole language approach, and we should have been using a phonics approach. And it is now really well understood. Lots and lots of studies have come out and said we need to move back to phonics. And this is showing up everywhere inside of the data. So the fact that 87% states have gone backwards, in the last 10 to 15 years, everyone should take a pause and realize we have more technology. The amount of technology, the power of technology has been doubling roughly every 12 to 18 months remain inside of the classroom. We put more funding inside of the classroom. We put more content in front of the student than we ever have. Over the past 15 years, we have thrown technology and content and funding, and we've gone backwards. One state has gone forward, and it was Mississippi. They went in from 2017 to 2024 from 47th in the country to 7th. Now that also tells you it's not a socioeconomic because Mississippi has a foundationally very diverse population. And socioeconomically, it is not one of the wealthiest states inside of the United States. The fact that we've gone forward is actually attributed to professional development and that they've made a huge commitment around what's called the science of learning, the science of reading. And so how they were teaching and what they were teaching made more of a difference than the tools, the AI or any of the software that we throw at them. And so from 47th up to seventh by actually doing things the right way. And McGraw Hill, we have over 274,000 titles, and we order content in a pedagogical manner. And so when we take something like a subject like Algebra 2, it has 586 different learning objectives. That means there's 200 to the 586 potential knowledge states when a student walks in that we have to develop. And you don't just achieve competency in Algebra 2 by parking somebody in front of a stack of books or in front of an LLM. you actually have to have a pedagogy about how you teach each 1 of these things, how you spiral back and make sure that you're reinforcing particular concept when you recognize or weaken something, reinforcing it. With a tool like Alex, we do exactly that. We actually complement the teacher. We create pie charts, we show proficiency. This is what I mean about precision learning that understanding the 2 to the 586 learning states. And so it's not just like do you have the content. It's have you ordered it in a way that actually is going to land. The second thing, so this pedagogical content is essential, and we've got an unmatched amount of it and history of producing it. The second thing is our precision data. We have over 190 terabytes of just learning data alone. We get about 19 billion learning interactions with our tools per year. And so as we understand what keeps the student on the path or off the path, we probably have a deeper understanding than almost any organization in the world as it relates to some of these learning journeys. It's one of the reasons why we're so excited about Magraw Hill Plus, where we allow third-party assessments, and then we also have our own assessment. So we have an open data model to understand what we call the longitudinal learning record of the student. And we just like to have a longitudinal health record, we have this longitudinal learning record. And the last thing is we are innovating rapidly. We have put out over 12 tools just in the past 18 to 24 months alone, AI-driven tools. We have a tool called sharpen that allows you to have podcast of content. We can present with flash cards. I mentioned Alex Adventure is a great tool that we have for K through 5 students. And so the third leg in our stool is the idea of these personalized tools. And so between those 3, we have an extraordinary position in learning.

Phil, can I get you to maybe expand on that last point, right? So the AI tools, AI reader, rider teacher assistant sharpen maybe just expand a bit on how these are shaping kind of classroom behavior or purchasing decisions? And then should we think of these products as sort of creating a tangible revenue opportunity? Or is this more about kind of enhancing the core offering?

Each of these tools, it's really interesting. -- teachers, we just we're working with a school district in the Northwest. And one of the biggest focuses was whether or not there was more than 20 minutes of class time that was dedicated to screen time. And there is a significant focus right now around the balance between screen time and human time in the classroom. And I'm really proud of the way that we've struck this balance. We have a tool like Alex Adventure. It's gamified. You get coins, you're able to buy new clothing for your Avatar, you're able to decorate your space in your virtual space. And we gamify learning of mathematics. And we saw increases of as much as 25% in first graders and then 55% in second graders, with gamification of some of the math concepts. A tool like Sharpen, really excited. We had -- which is all the way up in higher ed. This is the tool I mentioned, imagine almost like Spotify for education. You can take any of our content. We have about 185 of our titles that are already in sharpen and the professor is able to have kind of a closed environment. So it doesn't hallucinate, I just actually just produces its output based on our content or the content of the professor. In that tool in the study, Rowan University, they had about 60% of the students actually use this tool. And for final exam grade -- final exam grades of the 60% went up by 47% In-class quiz grades went up by 23% and and final grades went up by 21%. Now again, this is a tool that has a closed loop. When we build, we build with -- not just with the student, so we're not just a consumer company, but we also build with the enterprise. The professor in mind. And my background is a lot of this is being able to take what were consumer products, Microsoft Windows at the time and turn it into an enterprise product. And the ability to be able to close the loop with Professor give a closed environment, a garden, if you will, to be able to say, what should be in the garden at this moment versus not for the professor and then be able to close a loop and be able to say this student is struggling with these flash cards, the students listening to this podcast multiple times. This student seems to be getting something wrong in this game. We have 1 professor that's taken our Alex map up in Pittsburgh and they take what we call the pie charts, and each student has their own pie chart, and they can watch it bloom as they grow in competency. And then you put all the pie charts on the screen. And at the very beginning, there's only 1 student that had kind of a proficiency level. By the end literally the entire gamified the classroom, every one of the charts was completely filled out. And so the exciting thing for us is that we are one of the few companies at scale that can do this across the languages, across the different cultures, across the different infrastructures of the environment to be able to land with a professor, land with a teacher, teach them how to use the tools in the classroom and also land with the student.

Great. Maybe just one more, sticking on the AI topic. So in your fortune article, you stated the industries of the next 50 years require skills we haven't thought of yet. So how does that first inform your view of the associated education opportunity. But on the flip side, how do you think about the long-term risk of AI automating certain cognitive tasks or compressing the life of some technical skills where you guys offer a curriculum currently?

Look, this is -- as humans we've learned to live with lots of machines. The generative pretrained tram former is a really beautiful algorithm, it's another machine. We've got synthetic biology coming at us. We've got nanotechnology. We've got quantum computer coming -- quantum computing coming at us. We have a lot -- we have autonomous vehicles coming at us. We assume they're going to have to live with lots of machines. It is essential that we build great intelligence to be able to manage these. When I look at what we also have to do from an education perspective is that we have to prepare the workforce of the future. Everything I just mentioned to you is multitrillion-dollar industries that have yet to really even start to come about. And so we have to build the intellect. And humans knowledge right now, I just read a paper that in computer science alone, roughly about every 9 months right now, our knowledge is doubling. And so humans have to keep up. We, right now, we're really proud we just released a K-12 AI curriculum. You're going to see us doing some pretty extraordinary things. We just -- there were some new nutritional guidance as it related to the medical profession. We just released new -- entirely new nutrition curriculum for the medical field. We view this as the market and the TAM is exploding in terms of humans -- the potential to be able to train people in new concepts. Very few people that I know sit down in front of an LLM today and learn. You don't learn Piano in front of an LLM today. You don't learn -- if you've got to go off and do advanced calculus, I haven't heard anyone yet that sat down and really mastered advanced calculus. Let alone some of the professions or some of the more professional fields. We still believe that humans, teachers and students are going to be in a classroom learning 5 years from now, 10 years from now. And we believe they're going to have to learn even more. We believe the medical profession is -- you have to jam more and more into that 4 years. We think there's going to be a really exciting opportunity to be able to rethink medical curriculum. And so all of those present amazing opportunities for us, I would say, in the next 5 years and quite frankly, over the next 100 years.

Got it. Let's pivot to our operations in K-12. So the selling cycle worked against the company in fiscal '26. But in February, McGraw Hill confirmed an expectation for a TAM increase this year. There are several changes that we should cover at the state level, but just more generally, can you update on your medium-term view of the K-12 TAM.

So I have to always start because June 11, I'm releasing earnings. So we have -- Danielle is looking. So sometime beyond the horizon, is what I would say to all. Look, we were very happy of our most recent quarterly results. We really have been continuing to see great activity going on inside of the higher ed space as evidenced by the previous quarter. There are a number of states, California, in particular, that has pulled forward a number of decisions. And that's actually helping to increase the near-term TAM. But more importantly, I would say, again, in the literacy space we're seeing a significant amount of activity around literacy. As you can imagine, post that New York Times article and just in general, what's going on, we're seeing so many school boards actually reassessing literacy. Literacy is 40% of our revenue in K-12, 40%. And so as you see things like literally becoming more of a focus as you're seeing more and more requirements in fields of science, as I mentioned, in the Middle East, we're required to actually have curriculum around artificial intelligence, but as you see more impact around sciences and more focus on sciences and social studies and mathematics. We're seeing that marketplace, I would say, be very vibrant.

Okay. Maybe just -- you did a great job of explaining the science of reading earlier. Maybe we could tie that specifically into kind of the California opportunity. And maybe just discuss how McGraw Hill is kind of positioned into that process?

So literacy, as mentioned, 40% of our revenue inside of K-12, -- we invested in a product previously called Wonders, -- it ended up being about a $1.5 billion product for us over the course of a little bit less than a decade. Emerge Summit and SOAR which is our brand-new product. We have seen very, very good efficacy from it and a lot of excitement in the market. I think I'm allowed to talk about because it is public, Seattle just selected us. This was back in the April time period. And so that's an urban district that is making a big shift inside of literacy, California similarly. It's one of the largest investments that we've ever made in McGraw Hill in curriculum. And so we're excited about what we're seeing in California. I'm excited about what we're seeing up in other districts as they're starting to move things forward. But we're seeing a lot, as I mentioned, a lot more districts starting to pull forward movement of decisions around literacy.

Great. And then maybe staying on California. So we estimate the current MAT cycle is the largest component of this year's K-12 TAM rebound. And I think you've noted publicly already some some early pilot wins here. There is a more complex procurement landscape to consider, though, there's more list of providers, some changing standards. Just how are you seeing that cycle currently? Any kind of risk of elongation to purchasing decisions?

In California, there have been -- and this comes back to the statement I made, there are more organizations bidding on California math than we've ever seen. And it's an extraordinarily competitive marketplace. And to the point where we had mentioned that there's over 2,400 tools in K-12, those were the incumbents. But we're seeing a lot of new organizations. And so I can't really comment on where we're at. It's a little bit too early right now. I've noted in the past that with that many participants, you're going to have some wins, you're going to have some not selected in the marketplace. But it's a little bit too early to tell right now.

Maybe just covering 1 of the large state, which is Texas. So there's a new adoption process there. It looks a bit different. There's annual purchasing. -- broader set of options. Can you just help investors understand some of the changes and how they should consider competition from open educational resources like Bluebonnet.

So open educational resources is very similar to open source in the software community. There's lots of wonderful businesses like every 1 of the cloud businesses have been built on top of open source software. Hadoop is probably the biggest Google Big Data is probably 1 of the best examples of an open source product becoming an extraordinary multibillion-dollar business. So that exists as well in the education industry. Open education resources is exactly that set of open curriculum resources that student -- or I'm sorry, teachers can use to be able to integrate, whether or not it's in K-12 or higher ed, these are open resources like an open curriculum. Texas has chosen to produce a concept that's called Bluebonnet which is the state has actually funded the creation of curriculum on top of open education resources. We spend hundreds of millions of dollars building curriculum. And we're kind of in that business. We spent -- I think we've heard that Texas has spent could be as much as like $80 million or so. We spent $67 million just on accessibility alone. And there are requirements that you must have accessibility. And so every curriculum now is going to require deep investment in accessibility. Bluebonnet, we're a partner with Texas. We have a lot of really great relationships and we're getting a lot of good wins inside of the Texas pace. We're happy with our relationships in Texas. Bluebonnet, generally, what we're seeing is that schools are trialing this with 1-year contracts. We generally are cutting 3, 5 in some cases, 9-year contracts. And so right now, they're trialing. A number of other states have looked at whether or not they want to be able to do something at Bluebonnet and recently, they've come out and definitively said no, they don't. When a couple of the states came out and said define know, it was like right after. It was right after it was discovered there were 4,000 errors in the Bluebonnet curriculum. As mentioned, we do a lot of work to curate and ensure accuracy and accessibility in our curriculum. And so when we look at Texas, we're going to be great partners with them. And I think that we're going to -- we'll -- they can work with our tools as well our software tools with their curriculum or they can work with us.

Great. So in March, you announced an integration with Renaissance to connect their Star assessments to your curriculum. Can you elaborate on the collaboration and discuss why, in general, you prefer to partner here rather than own that testing layer.

So proficiency in precision education, we think is essential. Data and analytics really should be -- should guide the pedagogical journey of a student. McGraw Hill Plus for us is that data layer. And so it's our longitudinal record of learning for a student. We've chosen to make it open. And we did that because we foundationally believe that there's lots of ways to assess a student. We believe that we've got some pretty good assessment tools, but you're going to see us just continue to be a great partner of organizations like Renaissance, Pearson, Curriculum Associates and WEA, all are feeding their content. We're able to take the assessment content. And so we -- over time, we look at that data layer as a great opportunity for us.

Great. Maybe shifting to higher ed. So Ed has been a standout performer for Grab this past year. Maybe can you impact the recent growth algorithm between factors like inclusive access, pricing, market share gains, enrollment?

So I think that in higher ed, there is this great conception that textbooks have become outrageously, and I mean outrageously expensive. They were when I was in college and the price kind of went up. But what we've seen happen over time, especially with our approach like inclusive access is that we brought that cost back down. And as a matter of fact, the average curriculum is roughly about $0.02 of the education dollar. So we're a very, very small component of the education dollar. Why do I say that? It's because we do think -- we very rarely hear pricing as a challenge. Now one of the reasons for that is while we were private, we did a lot of work to be able to get the right kind of model in place, the right kind of economic model in place. So that we can go down to community colleges and we could scale up to the largest universities in the world. And quite frankly, that we could go international where pricing is very varied. Inclusive Access allows us to bundle the price of digital access for a textbook or for content into the price of the college degree. And so today, roughly about out of the 3,500 universities in the United States, there's roughly about 2,000 that have inclusive access where they're rolling into the price of the college degree. Generally, this starts with 1 to 2 professors and quickly jumps up to 25, 30, 50 professors. We have an extraordinary penetration in the higher-end marketplace, where we're in as high as 80% of the universities with some type of curriculum. And generally, inclusive access is increasing at roughly about 100 universities or colleges per year. And so we think there's a lot of growth in that model. The other thing that we've innovated, which we're really excited about is our evergreen model. I have a really good friend who's a business professor. And he said to me a while back. He said, my biggest problem and using curriculum is that I don't have the most common case studies. I had some curriculum that I was using that was about hydrogen vehicles. He said in like the failure of hydrogen vehicles by a particular car manufacturer, we said, "I want to know about something today. And so professors and students the way to be able to drive engagement is having relevant current content. And that's what Evergreen is. We're literally releasing content as it relates to some of the most important fields, whether or not it's business, whether it's medicine, economics, a whole variety of fields, engineering, we're releasing content on a regular basis. And literally, it's just like an iPhone update, where I can adopt the curriculum, like and choose to adopt the new updates. And so as we think about the future of McGraw Hill, we think about this always on publishing of content, curated content that feathers cleanly into the syllabus that fits into the tools that the professor is comfortable with and that is priced just as part of the overall education dollar.

Maybe just following up, frequent inbound we sometimes get is how sustainable are the gains from inclusive access, given your penetration into schools or from market share, which I think has been hovering a little over 30% recently. Just how much runway do you see for each of these drivers?

We see a lot of runway. When we were sort of pre-COVID, we had about 20% market share, and now we're up to roughly 30% market share. That's primarily United States. I just got back from a trip down to Latin America. You've got a very vast and growing middle class in Latin America, very massive and growing middle class in the Middle East. We're excited about what we're seeing in those markets as well. In the United States, as I mentioned, landing with inclusive access just because we're at 200 of the 3,500, we still see great run rate -- runway, both with more universities and also with existing -- with professors that have not yet adopted in their classes. And so we really look at the opportunity both horizontal expansion and vertical expansion, both geographically and also in the United States.

Got it. And then maybe just going back to the pricing point. I think McGraw previously discussed executing that to inflationary levels. But how do you think about maintaining pricing power? And how does things like AI play a part as you put new features in. We?

Are -- as I mentioned, we took a lot of hard medicine as a company, and you have an innovator's dilemma situation when you make this kind of a pricing change over to just everyone has access because your bet is I can get more students and can get and I can be more relevant versus a higher price and fewer students because a lot of students just simply wouldn't buy the textbook. Or they didn't buy until they decide to drop out. And so you had this dynamic where you going to get returns. You have a whole variety of problems. And so we took some hard medicine to get the pricing right. And this -- I generally -- I was meeting with a community college, I walked in and I talked to them about the pricing of our Inclusive Access and they were like, well, this isn't not a problem for the student. It's really priced significantly below what it had been when you were shipping textbooks around the world. So we feel good about that. AI, as you can imagine, with my background, we're going to be using AI everywhere we can inside of the company be gain efficiency. We're really proud of the fact that we've got hundreds of software developers that are already developing with it. We think that it's going to allow us to produce more content to keep pace with human knowledge. We think it's going to allow us to do better accuracy checks. We think it's going to allow us to produce new tools. As I mentioned, we've got a couple of our tools where we've just launched them in the past 12 to 18 months. They jumped up to 1 million users. We're at roughly about 65% of all of our users use our relationships or digital relationships. In the higher ed space, obviously, it's much higher in the K-12, we still have to ship textbooks and workbooks, but AI is really going to be an accelerator for us in our content tool creation.

So we spend most of our time on K-12, a little bit on higher ed. As it relates to professional international, though, it'd be great to hear where you're most focused, where you see the opportunities?

So there's 18 million open health care jobs right now. We have in the high 80s percent penetration in medical schools today of our medical curriculum. If you're not familiar with a textbook called Harrisons, it's one of like the bibles in the medical industry with just a who's who, it's about a 4,000 page book of medicine. Most doctors still have it on their shelf, really excited about that. We have Boards & Beyond, which is one of the most watched sets to prepare you. It's 800-plus videos. My son just graduated medical school a little bit over a year ago. He was like, "Oh my God." He lived in Boards and Beyond as he prepped for medical. And then we also have some pretty extraordinary things like we've access medicine. And one of our excess medicine professors has millions of followers. She's an internist and she does prostrate cancer podcast, and it drives a huge amount of efficacy of our access medicine. And so we're really experimenting and not just experimenting. We're running a business that has videos, it has podcasts, that has deep content, respected content, and we're doing some really exciting things. Some of the -- our customers for that technology are not medical schools. They're actually pharmaceutical companies, they're hospitals. And so when we look at that field, we really think we come from this amazing position as 1 of the most respected providers of curriculum and we see an opportunity, again, to go horizontal internationally. As an example, there's a number of large countries that are making big decisions around medical and then also vertically to be able to go into more and more specialties in that space. And so we're excited about that. And as I mentioned, Latin America, we're seeing some really great wins in Latin America just across the board, and we've got a great set of relationships in Latin America. And certainly, in the Middle East, we're also seeing some really exciting -- we've got some great relationships as we've talked about in the past, with Kingdom of Saudi Arabia with United Arab Emirates. And we're doing some cutting-edge things. As I mentioned, one of the drivers of us releasing an artificial intelligence curriculum was that the Middle East, UAE and KSA are asking for AI curriculum from K-12. And you want to be in those places where middle -- we're basically -- the middle is really growing in the economy because the very first thing you do is you start to get yourself into the middle class is you either educate yourself or you educate your children. And we're seeing that dynamic really take off.

All right. We are exactly out of time. Thank you.

Thank you.