Episode Transcript
[00:00:02] Speaker A: The companies that will be successful in the future, not just in the localization space, but in the every market, in every industry, in every geography, will be the companies that create the frameworks and opportunities for their people to lean in, embrace AI, leverage AI and manage their own destiny as it relates to know their work and what they're doing. Companies of the future will lead AI, not be told what to do by AI.
[00:00:38] Speaker B: Hello, my name is Eddie Arrieta. I'm the Chief Executive Officer of Multilingual Media. And today in Localization. Today we have the pleasure to have a conversation with Will Rolands-Rees, who is the new Chief Artificial Intelligence Officer at Lionbridge. Will, welcome.
[00:01:05] Speaker A: Well, Eddie, thank you so much for having me. It's a real pleasure to be here with Multilingual Media today.
[00:01:12] Speaker B: That's fantastic. And we had our conversation off the mic and I'm really excited to kind of like get into it. Congratulations, Chief Artificial intelligence officers. And what do chief artificial intelligence officers do these days in language, the language industry?
I'm assuming it's not the same that they were doing two years ago, five years ago. I don't even think it existed a couple of years ago. Tell us a wheel everything about it.
[00:01:42] Speaker A: Well, I think you're right, Eddie. I don't know that this role really existed two years ago, let alone five years ago.
[00:01:52] Speaker C: Right.
[00:01:53] Speaker A: In the industry.
You know, when I think about what does a chief AI officer do in a way, particularly in a language company, in a way one could say they almost do everything right now.
[00:02:09] Speaker C: Right.
[00:02:10] Speaker A: You can't get away from any conversation with a customer, with the community of partners, with investors, with anybody in the language industry without talking about AI in some. Some context or AI dominating that AI dominating that conversation.
But I would say that I would boil down some of the things that I think about within the organization into maybe three or four, you know, kind of buckets of three or four things that sort of keep me up at night.
One of them is Lionbridge has sort of more recently this year, re entered the AI data training space. So that was as that was a business that we were in just three and a half, three, almost four years ago that we divested to tell us.
And given everything that's gone on with large language models and the evolution of large language models and the opportunity set and frankly the demand from many of our customers, big and small, we've reentered that space. So I'm leading that reentry, leading that business vertical.
One of my colleagues that you spoke with recently, Susan Morgan, is a partner in Crime for me. She leads sales for AI data training, and we work very closely together to sort of build and grow that business. And we can talk more about that later if interested.
The second thing is sort of think strategically about AI and how it impacts the language space more generally. And so there are some, some of our deep domain experts that are deep in the, in the weeds of sort of AI, but trying to think, how is this going to impact? How should we be thinking about things like commercial models? How should we be thinking about deployment and trying to, and trying to sort of think strategically about, you know, what's the role to play for AI within, within a language organization? And, you know, in many respects, that journey is never, never going to end. But in many respects, I think we've, we've come a long way and in my opinion, from what I've seen, you know, we're sort of leading the industry and in our vision and utilization of AI. The third thing, right, and then you could say it's the third and fourth is what, what we. One thing we believe really deeply at Lionbridge is that not only is AI here to stay in a critical component of a language company, but if you want to be a language company in the future, your employee base, your, your talent, your people, the passionate linguists that you have, they also have to be very AI savvy. And that, you know, one thing that we've been leading this year has been a real drive to get everyone in our company an AI expert, right? We've, everyone has a, a goal this year that they have to use prompting, right, and become somewhat of a prompt expert.
And what we found has, you know, what we found is actually the results have blown us away.
We have more than three and a half thousand people using our sandboxes weekly. We have found that we've created this unstoppable AI momentum within the company where we're looking at processes, we're looking at commercial offerings, we're looking at very small things and very big things, all related to delivery for our customers and the way that we run our business.
And we're finding that people are leveraging AI to go and drive efficiency, drive improvement, deliver better outcomes. And so we've been really inspired as a leadership team with the way that, you know, the Lionbridge Pride has embraced AI. And, you know, sort of part of my role at the beginning of the year in particular was how do we sort of set that up and get that ball rolling? And then. And now that's a, that's a pull. That is an Unstoppable force multiplier, we think for us today and in the future in the language space.
[00:06:50] Speaker B: This is already very inspiring. Will, I'm glad we're having this conversation. And before we dig a bit deeper into the long term impact, I want to go back to the conversation on what you were mentioning initially. So you were in data training before, you're now doubling down on data training.
Let's break it down a bit more into why are you doing that?
What's the strategy behind it? And of course I can make some assumptions because at every conversation I've already had on artificial intelligence, everyone says that every AA problem always will boil down to the training data. That's the phrase that I've heard repeated so many times. So it's kind of like, okay, if that's where the value is at, you're like, we're going to be right in the middle of it, what's behind it, and tell us more about it, of course.
[00:07:50] Speaker A: So look, Eddie, that's a great question, right? And thank you for that.
So I think about this in when I think about business and I've been in businesses before which we've trapped in other industries, other companies that we've either built new business lines or transformed business lines. One thing that always comes back to me is their customer demand, right? Are your customers asking you to get back into a space or to enter a space? And if you think about some of the characteristics of AI data training and building data sets, there are a couple of things that are really, really important. Whether you're a big company or a small company building a big data set or a small data set, one of them is that data sets require really large volume, right? So if you want to be successful in this space, you really have to be at a scale secondarily.
What we're hearing from our customers repeatedly and one of the reasons that we've been winning programs and winning work is this is an arms race for our customers, right? They are investing millions of dollars on the promise of AI and they know that there is a time to market revenue, efficiency, challenge. And so what they need is a vendor that they work with that's not actually a vendor, it's actually a partner that they work with that they can trust is going to deliver what you've agreed is going to be delivered. The scale, the predictability, the quality outcomes. And when I think about one of the things in our localization business and our language AI business that Lionbridge is known for, it's really about our quality of delivery.
[00:09:56] Speaker C: Right.
[00:09:56] Speaker A: And our program management function, and that's why we've been trusted for over 25 years by many of the world's leading brands to deliver for them, is because we know how to make it work, we know how to get it done. So that's really important. And then the third thing I sort of think about is it's not just about big data sets, it's actually about having global reach and really access to a truly multilingual crowd of people that are deep domain experts. And so what's interesting for us in this instance is I think we're one of the few companies that can tell you that we can find you linguists in pretty much every language in the world. Right? But not just linguists, people that might be passionate or interested in the domain in which you're creating a training data set for.
And it's that access to that global crowd and that global content, which is highly synergistic with the work we do in the language side. And so we see this world as people need big data sets, so they need to work with scale players.
They want, they need to get predictable deliveries, right? So because their time to market is really important, they're in an arms race. And a big foundation for that is access to a global crowd at scale. And those are things that we're really good at, right? And what we really, we're really well known for.
And if you take all of that, that's a good reason to get back into this space, irrespective, right. Our customers have asked us to get back in and we have. But there's also, if you think about what's happened with AI over the last three years, it used to be the purview of the companies that had the infrastructure and the machine learning expertise to be able to build those models. There was a scarcity of technology resources, right? Availability of compute power.
Well, Microsoft, Amazon, Google, et cetera, have been making cloud computing accessible to people at huge scale over the last three to four years. And what that's evolved is that that's created the large language model space that's made AI accessible not just to the biggest tech companies, but to everybody. And so what we're seeing is that AI is no longer this sort of tech problem, it's an everybody problem. And, you know, we're seeing a lot of opportunity to help people with prompt engineering, to help people with LLM output validation.
Because I think one of the things that's sort of sometimes missed in this AI training space is it's not just about Creating the right data. You've also got to test that it works on the other side. And you still, you still need a crowd to be able to, you know, crowd of people, domain experts, passionate people to be able to do that. So all of those ingredients have kind of come together for us and we think we're really well served to, to play in this space. It's high velocity, it's moving really fast.
Initial results have been really good for us and we're doing this sort of wide variety of projects that are really, really very interesting.
[00:13:18] Speaker B: What go, what goes into data training?
So, so if I'm a company and I come to you, what happens? What do I usually want from, from Lionbridge?
[00:13:30] Speaker C: Wow.
[00:13:30] Speaker A: So, so, so that's a, you know, that's, you know that I, I would, I would say to Eddie, how long is a piece of string?
[00:13:38] Speaker C: Right.
[00:13:39] Speaker A: Because we have been asked to do some really, you know, pretty, pretty, pretty varied projects.
One thing, you know, so, so we're seeing quite a few companies come to us and what they're looking to do is to build a, either some form of speech recognition system.
So they're coming to us. Can you get me a thousand people in Thai, in Vietnamese and Japanese and French, German, Mandarin, Cantonese, whatever it may be, wide variety of languages and I need to record audio snippets so that I can go and build a foundational model that's going to help me with voice recognition. So we've done five, six, seven of those projects just in the last sort of three or four months. Really interesting.
[00:14:35] Speaker C: Right.
[00:14:37] Speaker A: We also have companies that are trying to do fine tune or rag pattern their LLMs. So they'll say, look, here's a prompt, here's the response I'm looking for. Can you help us validate and test that? A, is, is that a good prompt and a good response for my golden set? But now I want to go and run that against, you know, my system. I want to go and use variations of that prompt input. I want to see if I'm still getting the same output. So that kind of output validation.
We have a large, we work with a large technology company that wanted to create a really large data set very quickly for text data. Right. And they wanted to be able to sort of adapt, you know, if you put an input in. They wanted to be able to create some adaptation around, around that.
And that was hugely interesting. I mean it was a million plus sentences. Sorry, a million plus segments, five million sentences that we created in four weeks for them. And that's a foundational model that we expect to continue to build on. We did that in 12 languages for them. We have someone else that wants us to record people having a conversation, like you and I are having a conversation now, and measure the emotion that we have against a series of prompted questions and do that at scale, produce thousands of these videos and thousands of these interactions with studio quality, recording studio quality, audio annotation associated with that. And so what I think we're seeing, it's actually companies.
The creative juices of companies are really coming through. We're seeing big tech. There's quite a bit of focus on AR and VR. There's quite a bit of focus on audio and speech recognition systems. There's a lot of focus on text, there's a lot of focus on LLMs, there's a lot of focus on adapting existing content and annotating it to create a new content set. And that's something that we've been doing at scale as well. So the variety is really like. It's one of these spaces where actually everything is possible.
[00:17:03] Speaker C: Right.
[00:17:05] Speaker A: Look, we have someone that's reached out to us and they've been developing a widget that measures smell.
They want to know, can we provide people to test whether it's effective or not?
I mean, that's so far from language. Right? But also it's something that's really interesting and sort of right in our wheelhouse to be able to do so. In a way, we're limited only by your imagination for the projects we've been doing.
[00:17:36] Speaker B: That's a fantastic place to be in. That's a fantastic place to be in. I'm very excited. I'm very excited about the future and very excited to also hear that it was your customers that were asking for this. Repeated several times, like, you know, they told us to do this. So how did, how did they tell you? What sort of things were you getting from your customers that just made it really obvious to move in this direction.
[00:18:00] Speaker A: So I think one of the things that I find so interesting about this space, and again, one of the reasons why I think Lionbridge in particular has been successful, if I come back to some of the things that we're really known for, is often when you work with research scientists on a project, they have an idea like, I want to do this, but they're not quite sure whether it will work or not. And so a lot of the work we do is actually in deep partnership and evolves in some respect, in some cases daily, in terms of the instructions and the methods and what we're collecting because we're learning together about what's actually going to work, what doesn't work, how to do it in a way that really provides scale and the kind of the quality outcomes. And so I think, you know, like, customers, they've got some ideas about what they want, right? And some of it comes traditionally, like, hey, we've got an rfp. We need xyz, please bid on it. Sometimes it's a conversation with a customer where they're like, we got this idea.
Can we experiment? And then you experiment, and then you keep experimenting for three months, and you've realized you've done a really big project with them, but you've innovated and experimented along the way. And one of the things that we hear from our customers is when we know we need to experiment, we call Ironbridge, because we know that you guys are agile and nimble and you're interested, right? And you're not, hey, here's the set of instructions. And, you know, we can't, we can't evolve from that. It's much more. What have we learned? How do we do it differently? Great. Let's. Let's move forward and let's change and let's adapt. And that's something deep within our DNA, right? That sort of scale innovation is something which, you know, customers are really valuing. And again, many of these buyers are people that we've worked with on the localization side for a long time. So they're like. So they say, hey, great news, you're back.
Let's talk.
[00:20:07] Speaker C: Right?
[00:20:08] Speaker A: And, you know, you know, again, I, you know, what we find is that it's very tiring, Eddie, because this is moving really fast and the timelines are really fast, and it's exciting, right? I can't imagine anywhere more exciting, to be honest with you, in the language space than where we are. Yeah.
[00:20:31] Speaker B: And again, it's a very critical change in perspective and pace.
We've recently named our December issue the most recent issue of Multilingual Magazine. We named it the Year of AI for the language Industry.
And you've mentioned the. One of those things that keep you up at night is the impact in the language space.
And when you think about, I'm going to put it in the terms of the humans, right, the linguists, the translators and the interpreters. I remember a day when I was a translator and I never was trained for it. It just stumbled upon it. I was not part of the industry. There's so many of those companies that I know in Latin America that don't have any idea what locworld is or GALA or any of these associations, yet they are part of the industry.
[00:21:27] Speaker C: Right.
[00:21:28] Speaker B: And there is, there is. There is an effect that now technology is bringing, and in particular, of course, artificial intelligence. In other conversations we've had here in localization today, it has. We've talked a lot about empowering, amplifying the effect of humans and then also putting a different standard to who is a linguist standard and who is involved in projects. So I'd love to hear your perspective on the impact AI has had in the past year. And Lion Bridge, of course, it's very evident, but in the industry as a whole. And what impact do you think is going to continue having in the upcoming months and years? Of course.
[00:22:06] Speaker A: Yeah. So look, I think that AI is and will continue to have a big impact on, you know, the linguists and the language community, sort of in general. Right. I think, you know, sort of three things, though, that I would, you know, three things that I think about related to this, though. Firstly, this is not the first big change for linguists in the localization industry. You know, translation memories came along in the 90s, right. Quite impactful. Right. Suddenly you didn't have to retranslate everything all over again. You could just translate. What changed then? Machine translation and neuromachine translation came along in the 2000s, and now we're in this sort of era of sort of AI.
[00:22:56] Speaker C: Right.
[00:22:56] Speaker A: And so this is actually an industry that has already experienced quite significant evolution. And you could argue that maybe the most impactful of all was actually translation memories back in the day, given the way that it changed the amount significantly, the amount of content that needed to be sort of retranslated every time.
What's persisted throughout that is that the language industry continues to boom and flourish.
[00:23:27] Speaker C: Right.
[00:23:28] Speaker A: Like every time one of these big changes comes along, there are people that put up the big, you know, doom and gloom sign.
I don't think it's doom and gloom at all.
[00:23:38] Speaker C: Right.
[00:23:39] Speaker A: No matter where you sit within that industry, people continue to create more and more content every single day.
[00:23:47] Speaker C: Right.
[00:23:49] Speaker A: One thing that Covid did is actually made the world a lot smaller.
[00:23:54] Speaker C: Right.
[00:23:55] Speaker A: And so people are looking to localize and create engaging experiences in more and more languages.
[00:24:03] Speaker C: Right.
[00:24:03] Speaker A: That's good news for, you know, sort of good news for linguists as well. So. So I think the first one is that, you know, AI is just the next stepping stone.
[00:24:12] Speaker C: Right.
[00:24:13] Speaker A: Sort of in that. Sort of in that evolution.
[00:24:15] Speaker C: Right.
[00:24:16] Speaker A: I think the second thing is when I think about when I think about this space is you can either fight it or you can adopt it, right? You can lean in, like across our approach as being lean in, right? And not just lean in, but sort of lean in and establish it as part of, part of your, you know, sort of a part of your DNA. You know, the, the, this sort of, you know, you know, I think sometimes people outside the industry sort of still have this image of translators. The sort of the Benedictine monk sitting there and reading, you know, the copy and then writing, you know, the other, you know, that sort of related copy, copy, right in, you know, from, from scratch. That's not the case. I mean, the translators have really been reviewers for a very long time, right. You know, with this sort of empty post edit process where they're reviewing and then edit content, you know, up for review. But it's just another sort of step in that process. So in some respects, sort of, you know, it's not, you know, sort of not, not, not changing. But I also think, you know, like, there's the reasons where people have needed human translators and wanted human translators in the past. Really high quality health and safety requirements. For instance, really high brand requirements where, you know, if you get the dosage, mistranslate the dosage for a drug, you're going to have a really material impact. If you mistranslate the instructions for how to fix the engine on an airplane, I'm using, you know, very basic, then, then you're going to have a health and safety reason. Or if you mistranslate, say an annual report for your company and therefore you create yourself a brand problem, you're still going to need translators heavily involved in that process. But also you're going to need translators and there's going to be opportunities for people to help people with LLM review with multilingual prompt engineering, right? We're bringing their expertise to something else that's related to the AI industry. And we see a lot of demand for that and we see a lot of our community and a lot of our freelancers leaning in, leaning in on those opportunities because in a way that's the next set of, that's the next sort of revenue for, for them and earning for them.
[00:26:46] Speaker B: And I think, I think the conversation has completely shifted in the industry from like what it was last year at the end of like the conference season and what it was this year. Many rebrandings, many restructuring and strategizing. And what I see is more efficiency. I see more efficiency. I see more respect to different crafts and sides of kind of like the Spectrums. That's really, really interesting. And positioning an organization as leaders in AI, it's a huge task and it has happened in 2024. It's definitely happened for Lion Bridge. Earlier this year you launched Aurora AI Studio. What better way to talk about expertise?
What can you tell us about Aurora AI Studio and what has happened ever since it was launched?
[00:27:39] Speaker A: Sure. So, so Aurora AI Studio has been a, been a great technology release for us. And you know, in a way what we did is we had actually this really cool crowd platform in our games division. So one of the things that you know like that I think is really interesting about Lionbridge is we operate in a number of divisions. One of them is our games division which has been know a great business for us, growing really fast. And they did an acquisition a number of years ago for a business called Game Tester and Game Tester was really about how do I get kind of be, you know, a SaaS platform for the game game companies to test parts of their releases with a crowd of, you know, a predictable payment crowd model. On the other side of it they had, you know, hundreds of thousands of people in that community.
We adapted. We've been adapting that for the, for the AI training space. So some of the needs are a little bit different. Right. In the, in the game space you tend to have a fixed price per tester, right? Like hey, I need a tester. It's the same rate. We have different needs for that. Like testers in different locales or prompt engineers or annotators in different locations in different markets need different rates. We've been adapting it to sort of support that, but kind of the premise is pretty similar like how do I create a technology enabler to enable us? You know if you go back Eddie, to what I said before, companies are looking for predictable outcomes. The program around programs making sure they're delivered correctly. They're looking for really global scale and they're looking for speed.
[00:29:29] Speaker C: Right.
[00:29:29] Speaker A: Because time is very much money for our customers. And so Aurora Studio has been a great enabler for us to do that.
[00:29:35] Speaker C: Right.
[00:29:36] Speaker A: Like all good software products, we've got a long roadmap of things we want to add into that. But as a capability for our operational teams to use to deliver it's been fantastic. And we've started marketing it and selling it to a number of our customers for them to upload their own tasks for the crowd in a, in a sort of a non managed service way. So they, they will, they manage the tasks.
We're just the software that facilitates that. And the early pilots on that have been, you know, really encouraging for us.
[00:30:11] Speaker B: Will, how does, of course, just shifting gears here a little bit, how does someone, with your profile, how does someone end up being at a role that didn't exist two years ago? How did, how did the role happen to be? How do you, how did your career evolve into, into getting an interest in this conversation?
[00:30:34] Speaker A: Wow. Sometimes that's a great question, Eddie, because sometimes I don't know the answer to that.
[00:30:39] Speaker C: Right.
[00:30:40] Speaker A: And I've been asked that by, by, I've been asked that by, by friends, I've been asked that by peers, I've been asked that by mentors. Like, how did you end up in that role?
I've been asked that by my kids.
So I might answer it slightly differently than the way that you asked it, if that's all right. Which is what do I think makes a good chief AI officer? Right, Because I think that is.
Or what do I think some of the characteristics of a chief AI officer need to have? And what are some of the things I try and bring to the role that I, I think my experience sort of, I, I think my experience talks to?
Firstly, you have to be interested in the space, right? And I think that there are companies not in the language space per se, that have a chief AI officer that is the person leading AI development, right? They're a machine learning expert. They're an AI expert, right?
That's one role. The other side of AI is how do you embed AI within an organization? How do you go and solution with your customers, AI based, help them on their AI journey? And that's the side of, you know, sort of chief AI officer that I play. But the first one is you have to be really, you have to be curious about the space, you have to be knowledgeable, you have to be interested about what's going on. The second one is, and, and this is something which I have a lot, you know, quite a bit in my background. I think you have to know how to run a business because this is about, you know, at least the way that I've looked at this is you, you're thinking about this as a, essentially a business role, right? Now I have the AI data training business that we talked about, right, which is a direct P and L that I manage. But you know, this is the other elements to it. But you have to be thinking about this from a business perspective, not purely a technology perspective, right? Because you have to think, how do I use this to make money, right? At the end of the day that's the most important thing for any company is how do I make money? Whether that is in my role in this internally within Lionbridge, or when I work with our customers and talk to them about their AI journey and give them advice or solution with them, I have to know how to make money for companies, right? And so I have a background in running businesses. I've run portfolio companies for a very long time within my career. So I think that's, so I think that's, I think that's, I think that's a sort of a key, a key tenet.
The last one I would say is that you have to know how to work in a matrix and influence and you have to have some familiarity with transformational change because AI, as we've been discussing, is a major transformational event for the localization industry. So if you've not done transformation before, if you've not transformed a company, if you've not been within a company that's gone through a big change or something really sort of existential like, like AI is on, on the localization space, it's going to be very hard to bring some of those learnings and experiences to bear on this role. But also you're expected, the outcomes are big, but you're, you need to work with your peers across the company, right? You need to be able to influence and lead through influence.
And, and I think, you know, like some of the, some of the companies I've worked in, matrix leadership has been a really, really important part of their DNA. Like I know some companies like matrix, some like command and control. The companies I've worked in generally have been highly matrixed. And I think that's really important if you want to sort of leverage a role like this and leverage that in a transformational way to be able to influence the kind of change that not just is coming from outside, but that you want to embody within your company. You've got to, your peers have got to be on that journey. The levels down, you have to be at a. You have to be able to influence through, you know, across a matrix, as opposed to direct command and control. I think to be successful. So, you know, when I think about what this role needs, right, And I know I haven't directly answered your question, Eddie, but I tried to answer in a different way. I think you need to know about business because I think ultimately AI is a business outcome for a company needs how to make money, how to make more money, how to support your customers, how to advise them. I think that's really important. Need to be able to work across a matrix. And frankly, you need to be curious and interested.
[00:35:27] Speaker C: Right.
[00:35:28] Speaker A: And so, you know, John, our CEO, thought that I had some or all of those characteristics, which is why he asked me to take on this role.
[00:35:38] Speaker B: And it's really great to see because it's more on what you're made of rather than what are the specific tasks that got you here, the specific experiences, but kind of like your approach to it, which very likely will bring tons of innovation. So congratulations again. And I can definitely. It definitely resonates with me what you were mentioning about envisioning AI. It's almost like, like AI now. Yes. But technology as a whole as a. You find better ways of using not only AI, but other tools around. And you're right. As a business owner, what you're looking for is. And just as a note, I am also a bakery owner. So, like, I buy machinery so that the bakers get less tired. It's like, I don't want to. I don't want to get rid of you.
Never. I just, I just, you know, I want you to be less tired, more creative, more creativity and more innovation that come out of it. I know there are several of these cultural approaches within Lion Bridge. So from what you're doing and you know, what the culture is right now at Lionbridge, how is the team embracing technology? How's the team embracing artificial intelligence? And what role do you play there to influence it right now?
[00:36:58] Speaker A: Wow. So, you know, sort of interesting. I think when we started the year, John finally did one thing which I think was incredibly smart, which he met. He gave everyone an AI goal in the company and said everyone needs to have a goal that relates to learning how to do prompt prompting, something related to either making your job easier or helping us make more money. I'm not going to be more prescriptive than that, but everyone needs an AI goal in the company.
So it was set right from the top of the company and it was set in January.
And he's been very persistent every single week talking to people about, you know, sort of talking to people about that.
You know, when you start a goal like that, you're not sure how people will embrace it initially. And what we try to do is to give people a couple of. A couple of things that would enable them to be successful.
[00:37:58] Speaker C: Right.
[00:37:58] Speaker A: If they then chose to lean in. One of them is we created a sandbox environment using our, you know, we're a Microsoft Azure shop. So we used, you know, Azure GPT, right. And created an Azure Sandbox and made that available to everyone internally.
And then we also gave people training on how to use, how to prompt.
[00:38:25] Speaker C: Right.
[00:38:25] Speaker A: And that's partly what content not to put into the prompt.
[00:38:31] Speaker C: Right.
[00:38:31] Speaker A: So it's trying to give people the frameworks for how to safely use prompting, but also some of the skills and examples to do it. And then the third thing we did is we kicked off a series of showcases and every single month we would invite leaders from around the company to bring their team members in that had done something innovative around AI that they could showcase to everyone in the company. And the entire management team attends every single month. In fact, we've got our next one is tomorrow, our final one of the year.
And these have been a runaway success, right? I mean, if we run it to two hours, the hardest part of that is cutting down the number of ideas that people that want to present versus trying to find ideas. And we also ran an innovation contest where we would rate, we would ask people to submit their ideas and then we would give a cash prize out to people, which it was quite a meaningful cash prize.
It was a thousand dollars, couple of thousand dollars for winning, which depending on where you are in the world, is quite material. And we didn't geographically adjust it. So if you won and you were based in our China team or India team, you got the same prizes if you did in the United States. And that was also being a great motivator. And so what styles, if you did in the United States, and that was also been a great motivator. And so what started at the beginning of the year, in a way as a hope and an experiment by March, was a full blown movement that was unstoppable for us. And now as leaders, our role in a way is to continue to provide that kind of safety in the frameworks and frankly, get out of the way of our amazing people around the world that are innovating every day, leveraging AI in ways that they never thought they would be able to do, and solving real world business problems for us that impact our ability to run our business and operate it and also deliver for our customers in ways that, you know, we could do before, but might be highly manual, highly inefficient, or create new insights along the way that allow us to do things better and differently. And so frankly, it's been inspirational, right? And seeing people from the company post on LinkedIn that, Hey, I was just glad to present at this thing. I just won this award for my AI ideas, hugely inspirational for me. And I know the rest of the leadership team.
And I think one of the reasons it's being successful and I've seen other companies try different approaches, but I always think there's two approaches that you can take to something like embedding AI within your company. One of them is you can be very prescriptive from the top down and say solve these problems right, or you can say here's the framework, here's how I would like, here's what you can and can't do.
But you on the coal face of the day to day problems know better than I do what the problems that are worth solving it are. Don't solve those. And as you solve those, you'll wind up actually solving the big problems that as a management team we might have imagined were there and worth solving, but you'd have solved them organically from the bottom up. And that's I think why people are so inspired. Because I mean, I mean this like with all humility that I can muster because I'm really proud of this.
You know, people are really inspired within the company and really grateful that we've given this opportunity to go and make a difference and get recognized for it and be rewarded for it and please keep doing more because this is the future of where we're going. And people have felt really, really empowered and we've seen it embrace from delivery, from sales, from marketing, finance, hr in fact are one of our most innovative teams around the use of AI within the company and in some respects leading the way. So there's been a big, you know, this big movement.
We sort of planted the seed and you know, we've sat back and watch it grow very quickly into an unstoppable movement and we're really, really inspired by it.
[00:43:06] Speaker B: Yeah, and so will be our listeners and myself, that's for sure.
It's leaning in what you mention it into artificial intelligence. And I'm so glad we've had this conversation. Of course time flies when you are having fun.
We will definitely catch up again in the future if we coincide at an event for sure we'll have a conversation to see how things are going and evolving, which they will surely do.
Is there anything you'd like to mention to our audience, to those that are listening that we have not mentioned yet before we go?
[00:43:48] Speaker A: So, so I, I, you know, so firstly I just want to say thank you Eddie, for the opportunity to be here with multilingual and sharing some of my, sharing some thoughts and, and, and I, and I hope it, I hope there's something interesting and useful for Your listeners that the, the, the, that I've shared today, I firmly, I firmly believe that AI is here to stay.
And the companies that will be successful in the future, not just in the localization space, but in the. Every market, in every industry, in every geography, will be the companies that create the frameworks and opportunities for their people to lean in, embrace AI, leverage AI, and manage, you know, manage their own destiny as it relates to, you know, their work and what they're doing. Companies of the future will lead AI, not be told what to do by AI. And so I would just say three things in closing.
If your company hasn't leaned in on AI, okay, it's not too late, but it is kind of too late, right? But, you know, better late than never. Get on the bandwagon now. Figure out how to empower your people and figure out what your AI strategy is, because you need one no matter what your industry.
The second one is if you're dealing. If you're wondering how I should implement an AI system, build a foundational model and really good program management. Predictability of delivery is important. Getting access to a global, multilingual crowd is important. Working with someone that embraces change and innovation in the process and is going to be a partner. Shameless Plug, Please give us a call. We'd love to talk to you and we'd love to add you to our roster of clients.
And then the third thing is if you're a professional in the localization industry and you want to work for a company that is leading the way in AI, leading the way in localization with our Aurora platform, with the use of LLMs that we have, with a market leader in use of LLMs, it has a gains division which is just working with some incredible AAA publishers and titles. As an AI training business that is growing, an interpretation business that's growing. If you want to be part of that, whether you're in sales, operations, hr, finance, legal, technology, give us a call because Lionbridge is hiring and we're looking for people that embrace AI and want to be part of the future.
[00:46:48] Speaker B: So Shameless Plug, thank you so much, Will, that's fantastic. I'm sure a lot of our listeners will really appreciate all of the offers that you have made. It's been a pleasure getting to know you and we hope to have you here again in the future.
[00:47:04] Speaker A: Eddie, anytime. I'd be delighted to come back and as you said, hope to see you at a conference, but happy to do one of these Anytime that anytime that you think would be interesting.
[00:47:19] Speaker B: Fantastic. And this was our conversation with Will Rowlands Rees, who is the new Chief AI Officer at Lion Bridge. My name is Eddie Arrieta. I'm the CEO here at Multilingual Media and this was localization today. Until next time, goodbye. Goodbye, Will.
[00:47:38] Speaker A: Bye, Eddie.
