Exclusive Coverage: theCUBE at IBM Think | Dave Donahue & Emily Fontaine | IBM

theCUBE.net

Clips
News
More from Exclusive Coverage: theCUBE at IBM Think

David Donahue

Head of Strategy

unstructured.io

Emily Fontaine

IBM Office of the Chairman and CEO

IBM

Driving AI innovation: IBM’s strategic investments and startup collaborations

From established global giants and hyperscalers to nimble startups, accelerators and incubators, artificial intelligence innovation is flowing in from multiple channels.It’s imperative to give nascent AI innovators the needed support and collaboration to scale operational hurdles and concretize ideas. That’s the ethos behind IBM Corp.’s strategy.Discussing IBM’s AI fund and its support for startups such as Unstructured.“We launched the AI fund to invest in a range of companies from early to growth-stage startups,” said Emily Fontaine (pictured, right), Office of the Chairman and CEO at IBM Corp

play_circle_outline Unstructured's mission to structure unstructured data and its open-source library

play_circle_outline Importance of strategic partnerships and integration into Watsonx

play_circle_outline Focus on exceptional teams and market-critical solutions

play_circle_outline Unveiling the Power of a Strategic Alliance: Driving AI Innovation and Delivering Transformative Technologies to Clients

Info

Dave Donahue & Emily Fontaine | IBM

David Donahue

Head of Strategy unstructured.io

Emily Fontaine

IBM Office of the Chairman and CEO IBM

search

>> Hi, everybody. Welcome back to IBM Think 2024. We're here at the BECC in Boston where theCUBE got started in 2010. We're really excited to have Emily Fontaine, who's the head of IBM Global Ventures and David Donahue, who's the head of strategy at Unstructured. We're going to talk money, we're going to talk startups, we're going to talk innovation. Guys, welcome to theCUBE. Thanks for coming on.

>> Thank you for having us. We're thrilled to be here.

>> I appreciate it. Yeah, pumped.

>> Yeah.

>> So, IBM has a long history of M&A, investing in companies. Tell us about IBM Global Ventures. You've obviously got a fund, you've made an investment. We're going to get into it. What's the overall story of IBM Global Ventures?

>> Yeah. Well, thank you for asking. And as you said, IBM has a long history of AI capacity and delivering those to our clients globally and on the most complex challenges. We launched the AI fund to invest in a range of companies from early to growth-stage startups. Our goal there is to accelerate innovation at the infrastructure and enterprise application layers. So, we're investing in companies that have cutting edge solutions and that are bringing more robust, comprehensive set of capabilities to our clients. The last thing I'll say on this is our portfolio companies provide emerging market trends and deliver incredible value. In turn, we're helping those companies get onto enterprise ecosystems, and as a result, we're delivering our clients and partners best-of-breed, trusted solutions, they're getting after their most challenging problems.

>> So, of course, with a strategic investor like IBM, you're looking for a fit within the IBM ecosystem, within the IBM portfolio. And typically, you're saying early to mid-stage companies that have not necessarily scaled to go to market, is that right?

>> It really depends and it depends on the strategic fit, and where we're trying to go into that market. So, we made a huge investment in Hugging Face recently, which was a later-stage fund because of the strategic value to Watsonx, and Watsonx is all about what we're here to deliver gen AI to our clients. So, it really depends on the partnerships that we're creating and the value we can drive for our clients.

>> Got it. All right. Dave, I know you're not a founder, but put your founder hat on.

>> I'll do My best.

>> Why was the company started? I'll be interested in why you joined it, but what your role is? Yeah, why you joined the company? , you start the company, what problems are you guys solving?

>> Yeah, sure. So, the company's called Unstructured, right? So, I think the name kind of speaks for itself. We want to structure unstructured data. Unstructured data has been historically a huge problem for data engineers to solve, creating bespoke pipelines for downstream applications. And so, Brian, one of the founders, CEO and a couple others were saying, "This is crazy how much time and energy and professional services, companies that are taking to kind of solve these kind one-off solutions. Why isn't there just an agnostic platform that can take in any file type, any file format and transform that into a normalized JSON or normalized document?" So, that's what they built. So, it started with an open source library about a year ago. We've had 9 million downloads of that globally, which kind of gave us conviction around product market fit. And so, some of these companies, IBM included, there's probably 250 of the Fortune 500 companies using us. And a lot of those are kind of production-like workloads. So, some type of ongoing usage case, which suggests to us that there's a good path forward for unstructured and AI stuff moving forward.

>> Well, you asked a rhetorical question, why won't somebody do that? Because it's hard.

>> It is the most boring, difficult, time-consuming work. And so we're like, "Hey, nobody else wants to come after this because of how painful it is," and so we wanted to chase it. So, we've been kind of maniacally focused on creating solutions for developers to enable human-generated data to connect to machines. And so, it's great timing obviously with OpenAI and ChatGPT and all these applications that depend on a lot of data. And so, we're like, "Hey, we'll help you with that data problem."

>> I just want to add here something.

>> Yeah, go ahead.

>> I think it's so interesting for me about Unstructured and why I was so excited they're a portfolio company is the fact that all of our clients come to us and they have massive amounts of data, and they need to be able to ingest that and figure out how best to deliver insights on that and make actuals and sort through that data. And Unstructured really is an opportunity for them to be able to do that in a meaningful way. And that's why our partnership is so exciting.

>> Yeah, so a lot of synergies within IBM and across the portfolio. So, as a capital allocator, a typical VC would say, "I'm looking for market size, team and potential for moat." How do you think about those in order of importance? Market size maybe isn't as important for you because you got the biggest market in the world, but I presume team and tech are important.

>> Yes, they are.

>> So, what was it about the team and the tech that interested you?

>> Yeah, the first thing I'll say is we have a differentiated model and we are very much looking for exceptional teams that are delivering market-critical solutions. That's the most important thing that we're doing. We use a series of experts across IBM. We also use external experts. We have a fantastic portfolio team with a ton of experience, and we're really investing in the current and future leaders of AI. They're going to help us unlock the potential for the businesses. So, that kind of off the bat.

>> When they invest in us, we had very little commercial revenue. So, we were just getting off the ground. So, part of about the team and the moat, I mean it's about our founder and our leadership team, but this is a huge problem space that IBM is going to help us kind of unpack, which is awesome.

>> And I think, Dave, to that, I think what's exciting for us is we've developed a meaningful partnership, which has been hugely important. We're helping you unlock doors to go into market opportunities, as well as products, engineering, and then all of those key interlocks with potential clients, clients that you may go after by yourself, but clients we're going after together and creating these partnerships, it's so invaluable.

>> Well, it's huge because startups are risky. And if you come in with an IBM... You don't have a board seat, correct?

>> We don't have a board seat.

>> And how much did you invest?

>> But we're an observer seat. And Dave, I'll let you answer that question if you want.

>> Yeah. So, I mean IBM is intimately involved in everything. We treat them just like we would... They're people with the seats. This is a partnership that we want to continue because we understand the value that they can help us bring to market and vice versa.

>> Okay. So, you had mentioned I think 9 million downloads you said, which is sort of an indicator of product market fit.

>> Correct. Yep.

>> Tell me more about that. I'm always asking startups, "How do you know when you have product market fit?" Because so often I'll see startups try to scale go to market before they have product market fit and it doesn't work, and they ended up firing all the sales people they just hired. Now, you've got to go-to-market partner, so it's a little different, but double-click on that.

>> Correct. Sure. Yeah. So, I think the 9 million downloads, you look across a bunch of companies that have open source Apache libraries or Apache license with libraries in the open source and GitHub. 9 million is a lot for this type of technology. So, it was a validation and the investors we got in our series A and series B saying like, "Hey, we also know that this is a problem. Yes, the market is validating it with the number of downloads you have, but we also kind of being intimately involved with and having companies that are touching the space, touching AI, understand how big of a problem unstructured data is in order to..." The future of AI requires good data. And the only way you have good data is by accessing the rest of the data that's unstructured, 80% of enterprise data.

>> So, what's the business model in terms of how do you make money?

>> That's a good question. We're working on it.

>> Fair enough.

>> So, we're dual-use technology. So, we are involved on the government side, so we have contracts over there with a number of different partners. And then, on the commercial side, we just launched our first API, our SaaS API back in January. So, you can process, transform your documents as you need to through our API, and we process on a compute based or consumption-based model there. And then, what we're moving to is we'll keep the API, but then we'll also have a platform which is a self-serve model. So, people can access it on a per-page basis or per-page pricing model. But I think where we're going to get a ton of value down the road is why we have people like IBM as partners here because there's integrations that will help us to really create some growth in terms of our revenue. If we integrate with something like Watson data, that would be an enormous value for us in terms of growing our visibility, but then also in terms of revenue down the road.

>> And I would just want to add, and that's something that, Dave, we've talked a lot about over the course of a ton of the last few days is the next step for us in our partnership is that integration into Watsonx. We're here at Think, it's all about Watsonx for us. And so, making sure that our portfolio companies, like Unstructured, are getting plugged in and they're integrating into Watsonx to deliver that extra value to our clients.

>> Yeah, yes.

>> Go ahead please.

>> Oh, I was going to say, so I mean right now, companies have access to... Historically, 20% of their data is structured. So, that's what they're accessing. And we're saying, "Hey, do you want to 4X that with the other 80% that we're enabling through our product to be able to feed into what the the Watson data lake or Watson data?"

>> Yep.

>> So, I'm interested in your philosophy. I'm going to ask you a weird question.

>> Okay, I love a weird question. Hit me with it.

>> In horse racing, , but I'll make it relevant. In horse racing, a favorite comes in about 33% of the time. But the public are actually really good handicappers because a two-to-one comes in more often than a three-to-one or a four-to-one. So, actually, the pattern is right. Why is that relevant? Because when you look at the data, the percent of startups that fail that are series A, B, and C is the same. It's like 70%. So, as an industry, we're horrible at making bets. How do you think about that? And as a startup, you want a VC that's founder-friendly, most aren't to be honest with you, but a strategic investor maybe is a little bit more founder-friendly.

>> Unstructured is very friendly. I mean, let's put a plug in for them.

>> So, I told you it was a weird question, but how do you think about that? Basically, what I consider a failure to allocate capital because it's almost like the industry doesn't care. As long as we get one out of 10 and it's 100 bagger, we're fine. But that's not startup-friendly. How do you look at startups?

>> I love this question and I think I touched on it a little bit more, but I want to be really clear on my answer. We are looking for exceptional teams and we very much over-index on that. And we are all about finding market-critical solutions. Solutions that are going to help elevate our clients, that are going to help fill gaps that we have that are going to bring new emerging trends, new insights. Because even if a company fails, it's not a failure to us at IBM because we are developing and understanding where the market is going, where the trends are going, what the insights are going. And so for us, as long as we're learning something, it's hugely invaluable. Additionally, we are making great investments, we're making in Unstructured. We're creating these and growing our ecosystem partnerships, which is incredibly invaluable to us. We are very much leaning in, as I said before, $500 million funds that we're all about here. It's not our only venture fund, but our most critical. And we are looking for strategic partnerships that are going to help us grow that. They're going to help us leverage Watsonx. They're going to integrate and leverage Watsonx. And they're going to help us continue to grow our long-term relations with our clients. We're IBM, we're all about clients, we're all about partnerships. Our venture fund is helping us unlock more potential and greater value for our clients.

>> Yeah, small lever can move a big rock.

>> Exactly.

>> And that's what I'm hearing here in terms of the philosophy. And startups drive innovation, right? I mean that's-

>> I'm a huge component of innovation. I'm always talking to Dave's ear off about this, but startups drive a ton of innovation because we connect our startups, our portfolio companies with all of our business units, our software, our hardware, our consulting teams, our research teams, they are getting a piece of all of IBM, which is hugely invaluable to them, but also to us because they're delivering, as I said, those insights, those market trends, and it's critical.

>> Yeah. And I think just to kind of talk about the partnership piece a bit. We were lucky as a startup raising money to have... There's a bunch of people who wanted to join in on the round, I guess for lack of better words. We had a choice in the partners. We chose IBM because we understood the value that they could provide to us. And you were talking about like 30% are successful, this is a bet on our part and kind of the strategic partners about who's going to make us successful down the road. And IBM is a massive global organization, but this partnership with steering us directly with the people that are going to be the most helpful for us in these integrations, in getting the product into customer's hands, into kind of the IBM architecture is so incredibly important because a startup would get lost here otherwise.

>> Yes, .

>> Sorry, can I just add one more thing?

>> Yeah, absolutely.

>> Okay. I just want to make two more points. So, one point I would make is once we invest, that's just the start. Our relationship is so much more to that. It's a go-to-market relationship. It is a helping connect with clients. We've done that a bunch today. We're connecting, as I said, with the business units. So, when we invest, we think about it as a long-term relationship in which we are going to go to market together. And so it's not just about... Initially, it starts with investing, but it comes much more than that.

>> How long have you been in this role?

>> I've only been in this role for a little bit. I've been at IBM for a long time.

>> Okay. So, my last question, and Dave, I'm going to start with you and then you can bring us home, Emily.

>> Yeah, please.

>> A year from now we'll be at Think 2025. What would you like to say, Dave, then that you can't say now?

>> What would I like to say then that I can't say now? We are just starting this relationship right now. I would love to say that IBM was the reason why we took off, why we are in everybody's architecture for enabling AI down for these AI applications downstream. And I think it started with this relationship, really with the investment. But I can't forecast the future, but I would love to have this be the start of something pretty awesome down the road.

>> Yep.

>> Great.

>> And for us, if I can add, we want to see Unstructured embedded in Watsonx. We want to see all of our clients using it and we want to get after clients together. And so, for us, I just want to say we are doubling down on our commitment for AI innovation, and we want to use companies like Unstructured to deliver transformational technologies to all of our clients and all of our partners.

>> All right. Well, I'll see you guys in a year.

>> Yeah, I would love that. We would love that.

>> And we'll have this conversation, right?

>> Yeah.

>> Really enjoyed the conversation.

>> Thank you so much.

>> Thank you so much and congratulations.

>> We appreciate you.

>> Thank you.

>> Okay. Keep it right there. We're here at the BECC in Boston, theCUBE, go to siliconangle.com, thecube.net for all the action. My name is David Vellante. We'll be right back.

Dave Donahue & Emily Fontaine | IBM

search

>> Hi, everybody. Welcome back to IBM Think 2024. We're here at the BECC in Boston where theCUBE got started in 2010. We're really excited to have Emily Fontaine, who's the head of IBM Global Ventures and David Donahue, who's the head of strategy at Unstructured. We're going to talk money, we're going to talk startups, we're going to talk innovation. Guys, welcome to theCUBE. Thanks for coming on.

>> Thank you for having us. We're thrilled to be here.

>> I appreciate it. Yeah, pumped.

>> Yeah.

>> So, IBM has a long history of M&A, investing in companies. Tell us about IBM Global Ventures. You've obviously got a fund, you've made an investment. We're going to get into it. What's the overall story of IBM Global Ventures?

>> Yeah. Well, thank you for asking. And as you said, IBM has a long history of AI capacity and delivering those to our clients globally and on the most complex challenges. We launched the AI fund to invest in a range of companies from early to growth-stage startups. Our goal there is to accelerate innovation at the infrastructure and enterprise application layers. So, we're investing in companies that have cutting edge solutions and that are bringing more robust, comprehensive set of capabilities to our clients. The last thing I'll say on this is our portfolio companies provide emerging market trends and deliver incredible value. In turn, we're helping those companies get onto enterprise ecosystems, and as a result, we're delivering our clients and partners best-of-breed, trusted solutions, they're getting after their most challenging problems.

>> So, of course, with a strategic investor like IBM, you're looking for a fit within the IBM ecosystem, within the IBM portfolio. And typically, you're saying early to mid-stage companies that have not necessarily scaled to go to market, is that right?

>> It really depends and it depends on the strategic fit, and where we're trying to go into that market. So, we made a huge investment in Hugging Face recently, which was a later-stage fund because of the strategic value to Watsonx, and Watsonx is all about what we're here to deliver gen AI to our clients. So, it really depends on the partnerships that we're creating and the value we can drive for our clients.

>> Got it. All right. Dave, I know you're not a founder, but put your founder hat on.

>> I'll do My best.

>> Why was the company started? I'll be interested in why you joined it, but what your role is? Yeah, why you joined the company? , you start the company, what problems are you guys solving?

>> Yeah, sure. So, the company's called Unstructured, right? So, I think the name kind of speaks for itself. We want to structure unstructured data. Unstructured data has been historically a huge problem for data engineers to solve, creating bespoke pipelines for downstream applications. And so, Brian, one of the founders, CEO and a couple others were saying, "This is crazy how much time and energy and professional services, companies that are taking to kind of solve these kind one-off solutions. Why isn't there just an agnostic platform that can take in any file type, any file format and transform that into a normalized JSON or normalized document?" So, that's what they built. So, it started with an open source library about a year ago. We've had 9 million downloads of that globally, which kind of gave us conviction around product market fit. And so, some of these companies, IBM included, there's probably 250 of the Fortune 500 companies using us. And a lot of those are kind of production-like workloads. So, some type of ongoing usage case, which suggests to us that there's a good path forward for unstructured and AI stuff moving forward.

>> Well, you asked a rhetorical question, why won't somebody do that? Because it's hard.

>> It is the most boring, difficult, time-consuming work. And so we're like, "Hey, nobody else wants to come after this because of how painful it is," and so we wanted to chase it. So, we've been kind of maniacally focused on creating solutions for developers to enable human-generated data to connect to machines. And so, it's great timing obviously with OpenAI and ChatGPT and all these applications that depend on a lot of data. And so, we're like, "Hey, we'll help you with that data problem."

>> I just want to add here something.

>> Yeah, go ahead.

>> I think it's so interesting for me about Unstructured and why I was so excited they're a portfolio company is the fact that all of our clients come to us and they have massive amounts of data, and they need to be able to ingest that and figure out how best to deliver insights on that and make actuals and sort through that data. And Unstructured really is an opportunity for them to be able to do that in a meaningful way. And that's why our partnership is so exciting.

>> Yeah, so a lot of synergies within IBM and across the portfolio. So, as a capital allocator, a typical VC would say, "I'm looking for market size, team and potential for moat." How do you think about those in order of importance? Market size maybe isn't as important for you because you got the biggest market in the world, but I presume team and tech are important.

>> Yes, they are.

>> So, what was it about the team and the tech that interested you?

>> Yeah, the first thing I'll say is we have a differentiated model and we are very much looking for exceptional teams that are delivering market-critical solutions. That's the most important thing that we're doing. We use a series of experts across IBM. We also use external experts. We have a fantastic portfolio team with a ton of experience, and we're really investing in the current and future leaders of AI. They're going to help us unlock the potential for the businesses. So, that kind of off the bat.

>> When they invest in us, we had very little commercial revenue. So, we were just getting off the ground. So, part of about the team and the moat, I mean it's about our founder and our leadership team, but this is a huge problem space that IBM is going to help us kind of unpack, which is awesome.

>> And I think, Dave, to that, I think what's exciting for us is we've developed a meaningful partnership, which has been hugely important. We're helping you unlock doors to go into market opportunities, as well as products, engineering, and then all of those key interlocks with potential clients, clients that you may go after by yourself, but clients we're going after together and creating these partnerships, it's so invaluable.

>> Well, it's huge because startups are risky. And if you come in with an IBM... You don't have a board seat, correct?

>> We don't have a board seat.

>> And how much did you invest?

>> But we're an observer seat. And Dave, I'll let you answer that question if you want.

>> Yeah. So, I mean IBM is intimately involved in everything. We treat them just like we would... They're people with the seats. This is a partnership that we want to continue because we understand the value that they can help us bring to market and vice versa.

>> Okay. So, you had mentioned I think 9 million downloads you said, which is sort of an indicator of product market fit.

>> Correct. Yep.

>> Tell me more about that. I'm always asking startups, "How do you know when you have product market fit?" Because so often I'll see startups try to scale go to market before they have product market fit and it doesn't work, and they ended up firing all the sales people they just hired. Now, you've got to go-to-market partner, so it's a little different, but double-click on that.

>> Correct. Sure. Yeah. So, I think the 9 million downloads, you look across a bunch of companies that have open source Apache libraries or Apache license with libraries in the open source and GitHub. 9 million is a lot for this type of technology. So, it was a validation and the investors we got in our series A and series B saying like, "Hey, we also know that this is a problem. Yes, the market is validating it with the number of downloads you have, but we also kind of being intimately involved with and having companies that are touching the space, touching AI, understand how big of a problem unstructured data is in order to..." The future of AI requires good data. And the only way you have good data is by accessing the rest of the data that's unstructured, 80% of enterprise data.

>> So, what's the business model in terms of how do you make money?

>> That's a good question. We're working on it.

>> Fair enough.

>> So, we're dual-use technology. So, we are involved on the government side, so we have contracts over there with a number of different partners. And then, on the commercial side, we just launched our first API, our SaaS API back in January. So, you can process, transform your documents as you need to through our API, and we process on a compute based or consumption-based model there. And then, what we're moving to is we'll keep the API, but then we'll also have a platform which is a self-serve model. So, people can access it on a per-page basis or per-page pricing model. But I think where we're going to get a ton of value down the road is why we have people like IBM as partners here because there's integrations that will help us to really create some growth in terms of our revenue. If we integrate with something like Watson data, that would be an enormous value for us in terms of growing our visibility, but then also in terms of revenue down the road.

>> And I would just want to add, and that's something that, Dave, we've talked a lot about over the course of a ton of the last few days is the next step for us in our partnership is that integration into Watsonx. We're here at Think, it's all about Watsonx for us. And so, making sure that our portfolio companies, like Unstructured, are getting plugged in and they're integrating into Watsonx to deliver that extra value to our clients.

>> Yeah, yes.

>> Go ahead please.

>> Oh, I was going to say, so I mean right now, companies have access to... Historically, 20% of their data is structured. So, that's what they're accessing. And we're saying, "Hey, do you want to 4X that with the other 80% that we're enabling through our product to be able to feed into what the the Watson data lake or Watson data?"

>> Yep.

>> So, I'm interested in your philosophy. I'm going to ask you a weird question.

>> Okay, I love a weird question. Hit me with it.

>> In horse racing, , but I'll make it relevant. In horse racing, a favorite comes in about 33% of the time. But the public are actually really good handicappers because a two-to-one comes in more often than a three-to-one or a four-to-one. So, actually, the pattern is right. Why is that relevant? Because when you look at the data, the percent of startups that fail that are series A, B, and C is the same. It's like 70%. So, as an industry, we're horrible at making bets. How do you think about that? And as a startup, you want a VC that's founder-friendly, most aren't to be honest with you, but a strategic investor maybe is a little bit more founder-friendly.

>> Unstructured is very friendly. I mean, let's put a plug in for them.

>> So, I told you it was a weird question, but how do you think about that? Basically, what I consider a failure to allocate capital because it's almost like the industry doesn't care. As long as we get one out of 10 and it's 100 bagger, we're fine. But that's not startup-friendly. How do you look at startups?

>> I love this question and I think I touched on it a little bit more, but I want to be really clear on my answer. We are looking for exceptional teams and we very much over-index on that. And we are all about finding market-critical solutions. Solutions that are going to help elevate our clients, that are going to help fill gaps that we have that are going to bring new emerging trends, new insights. Because even if a company fails, it's not a failure to us at IBM because we are developing and understanding where the market is going, where the trends are going, what the insights are going. And so for us, as long as we're learning something, it's hugely invaluable. Additionally, we are making great investments, we're making in Unstructured. We're creating these and growing our ecosystem partnerships, which is incredibly invaluable to us. We are very much leaning in, as I said before, $500 million funds that we're all about here. It's not our only venture fund, but our most critical. And we are looking for strategic partnerships that are going to help us grow that. They're going to help us leverage Watsonx. They're going to integrate and leverage Watsonx. And they're going to help us continue to grow our long-term relations with our clients. We're IBM, we're all about clients, we're all about partnerships. Our venture fund is helping us unlock more potential and greater value for our clients.

>> Yeah, small lever can move a big rock.

>> Exactly.

>> And that's what I'm hearing here in terms of the philosophy. And startups drive innovation, right? I mean that's-

>> I'm a huge component of innovation. I'm always talking to Dave's ear off about this, but startups drive a ton of innovation because we connect our startups, our portfolio companies with all of our business units, our software, our hardware, our consulting teams, our research teams, they are getting a piece of all of IBM, which is hugely invaluable to them, but also to us because they're delivering, as I said, those insights, those market trends, and it's critical.

>> Yeah. And I think just to kind of talk about the partnership piece a bit. We were lucky as a startup raising money to have... There's a bunch of people who wanted to join in on the round, I guess for lack of better words. We had a choice in the partners. We chose IBM because we understood the value that they could provide to us. And you were talking about like 30% are successful, this is a bet on our part and kind of the strategic partners about who's going to make us successful down the road. And IBM is a massive global organization, but this partnership with steering us directly with the people that are going to be the most helpful for us in these integrations, in getting the product into customer's hands, into kind of the IBM architecture is so incredibly important because a startup would get lost here otherwise.

>> Yes, .

>> Sorry, can I just add one more thing?

>> Yeah, absolutely.

>> Okay. I just want to make two more points. So, one point I would make is once we invest, that's just the start. Our relationship is so much more to that. It's a go-to-market relationship. It is a helping connect with clients. We've done that a bunch today. We're connecting, as I said, with the business units. So, when we invest, we think about it as a long-term relationship in which we are going to go to market together. And so it's not just about... Initially, it starts with investing, but it comes much more than that.

>> How long have you been in this role?

>> I've only been in this role for a little bit. I've been at IBM for a long time.

>> Okay. So, my last question, and Dave, I'm going to start with you and then you can bring us home, Emily.

>> Yeah, please.

>> A year from now we'll be at Think 2025. What would you like to say, Dave, then that you can't say now?

>> What would I like to say then that I can't say now? We are just starting this relationship right now. I would love to say that IBM was the reason why we took off, why we are in everybody's architecture for enabling AI down for these AI applications downstream. And I think it started with this relationship, really with the investment. But I can't forecast the future, but I would love to have this be the start of something pretty awesome down the road.

>> Yep.

>> Great.

>> And for us, if I can add, we want to see Unstructured embedded in Watsonx. We want to see all of our clients using it and we want to get after clients together. And so, for us, I just want to say we are doubling down on our commitment for AI innovation, and we want to use companies like Unstructured to deliver transformational technologies to all of our clients and all of our partners.

>> All right. Well, I'll see you guys in a year.

>> Yeah, I would love that. We would love that.

>> And we'll have this conversation, right?

>> Yeah.

>> Really enjoyed the conversation.

>> Thank you so much.

>> Thank you so much and congratulations.

>> We appreciate you.

>> Thank you.

>> Okay. Keep it right there. We're here at the BECC in Boston, theCUBE, go to siliconangle.com, thecube.net for all the action. My name is David Vellante. We'll be right back.