#249 - Generative AI in Recruitment: Bridging the Gap Between Automation and Authenticity

Speaker A: They say one of the best ways to further your career and get more impact in your organizations is by learning from people who have done it before. So people who are further ahead in the things that you want to do. For that, we at Data Futurology put together conferences, bring people together to have the conversations that matter to you with key leaders from around the country. Our next conference is called Opsworld. It focuses on building and deploying data products and ML products. This is happening in Melbourne on the 24th and 25th of October. Uh, we are having multiple industry use cases of data products and mill products. We have strategies on how to build those and additionally we have uh, lots of networking opportunities for you to spend time with CIOs, uh, CDOS, CDAOs who are coming to join us and present. Hope to see you there.

Speaker B: Click below for tickets.

Speaker A: I'd like to say a big thank you to our sponsors. Talent Insights Kelly Insights are Australia's leading specialist data recruitment business with offices in Sydney, Melbourne and Brisbane. They're experts at providing recruitment strategy and building data teams for clients across industries Australia wide. They provide recruitment solutions for all roles across the data lifecycle including data engineering, data science, advanced analytics, customer and marketing, Insights, business intelligence, data product managers and data, uh, governance. They're skilled at finding the best permanent and contract hires for your business needs as well as statement of work project focused data resources at ah, Talent Insights. Relationships matter most. I can say from firsthand experience, Talent Insights uh, are fantastic to work with. Whether you're a business leader within an HR network or a specialist data candidate. Talent Insights should be the first company you turn to for all your data recruitment needs. Find them@talentinsights.com hi, this is Felipe Flores.

Speaker B: Welcome to Data Futurology. Today we're going to talk about generative AI in recruitment, both from a uh, people perspective and a tech perspective. And for that I couldn't have asked for any better guests. Uh, I've got Grant Wright, uh, he is the EM of marketplace and AI products at Seek. Great to see you Grant. Good to see you mate. You're good. Thanks so much at the time. And I've got James Jimmy Acorn. He's the principal consultant for data engineering, machine learning, um, data science at Talent Insights Group. Jimmy, thanks for making the time. How you going?

Speaker C: Good, thank you. Thanks for having me along.

Speaker B: Uh, thank you both. So I'll ask you both to um, give us a quick intro and then we can jump into um, gen in recruitment. Uh, Jimmy, do you want to go first? Yeah, sure.

Speaker C: Uh, so I work for a business called Talent Insights, um, we're in Sydney and Victoria. We specialize in recruiting and data and analytics. So my area of expertise as you mentioned, is data engineering, uh, technical data science roles and machine learning.

Speaker B: Yeah, nice, nice that you were saying that there's been uh, a lot of interest in the MLOps space, uh, recently in data engineering. Is that the areas that you're seeing, uh, um, have a lot of interest and a lot of demand from the market?

Speaker C: Definitely, yeah. There's still a really big, uh, requirement to find people that have experience of, know productionizing models and having that really strong engineering skill set. So that's probably been a big focus in Victoria for the last, uh, last 12 months or so.

Speaker B: Yeah, I love it. And with uh, Data Featureology, we have our mll, our, our OPS World now OPS World conference coming up in Melbourne at the end of October, so October 24th, 25th. And it's focused on data products and ML products. Um, so it um, aligns well with, with where uh, the market's at, um, and Grant from your side mate, um, give us a bit of an intro.

Speaker D: Yeah. So I'm lucky to hit up AI products and analytics at Seek. So Seek's a global employment marketplace. Many in Australia and New Zealand will be familiar with that brand as the leading place to find a job, uh, in Australia and New Zealand. But some may not be aware of our presence in Asia Pacific in the Americas. So through the, uh, Jobstreet, JobsDB, OCC and Catho brands, uh, we've also got leading positions right across Asia Pacific, um, Mexico and Brazil. So my team's job is to use data and AI to make it easier to find the best job and the best talent, uh, on Seek, uh, globally.

Speaker B: Globally. That's amazing. Yeah. And um, the acquisitions that Seek has done over the years and um, overseas just I've given you guys such broad, um, data sets and uh, information and where you can apply your AI magic, uh, to help people find the best, the best jobs.

Speaker A: Uh, so.

Speaker B: That's amazing, mate. Um, so as a continuation of that, I might ask you, how long has Seek, uh, been using AI in production? What, uh, is, uh, an overview of the journey so far? Um, what does that look like?

Speaker D: Yeah, so seek's been using AI. I've been in seek about seven years. Seek's probably been using AI in anger in the marketplace for about 10. Um, so we started with a pretty small team building recommendations, algorithms and a few prediction models. Um, I joined the team about six years ago with the goal of Kind of making a global team to do things once globally. So we've been on a pretty strong growth trajectory there. We started with a couple of people, agreed to a data science team of about 20. Um, now we're in the hundreds with data scientists, product managers, engineers, uh, and platform teams. So uh, building about 75 services across 25 countries. Uh, if you include all our services in about five languages. So it's been quite a journey.

Speaker B: Mate. That is incredible. Yeah, so many different uh, areas of application. Um, so what are some of the examples uh, that you can share about AI products in Seek?

Speaker D: Yeah, we use AI almost behind everything you use on the Seek website. So everything from M helping you predict your search query through Autocomplete, right through to personalized recommendations being pushed to your mobile. Um, things that people wouldn't necessarily see day to day as a candidate searching the site. So also recommending candidates through your Seek profile to hire us when they post a job ad. Understanding, um, and extracting skills and trying to understand career paths so we can open up new opportunities and explain those to candidates and hires. Um, and also pricing and performance management of our products to make sure we're allowing hires to get the performance they need but also making sure we're always putting the most relevant thing in front of the candidate.

Speaker B: Right. That's brilliant. And um, how does your team, uh, what's the organization? I guess the structure of the team and how do you operate to get the maximum business impact?

Speaker D: Yeah, that's something we think a lot. So I'm not a data scientist. My background is kind of strategy and operations with an analytics bent. Um, and the reason I think that role, my role exists is because organizing highly capable technical teams around customer problems and giving them the context to go and solve, it's just so critical for what we do. Um, so we organize, we have platform teams which do support and excellence. So things like experimentation and data engineering to provide support to all our teams. Um, and then we organize into kind of customer domain teams which focus around a customer problem like search for example, um, job search. And those teams have AI product data science and engineering in them. So they're a cross functional team with slightly messy boundaries because we think the innovation happens between those three things. Um, and then we have pretty clear, we push pretty hard on having a clear product ambition and understanding what the real customer problem is we're trying to solve and a theory on how data and AI can unlock a constraint and make that dramatically better. And that's what the team chased and then they have a Fair bit of freedom to then come up with okrs, uh, underneath that, that align to that goal, uh, and tell us how they're going to solve the problem.

Speaker B: That's awesome. That is great. Um, and then jumping uh, more into the um, LLMs and Engine AI topic, um, first from a SIG perspective, have you guys been looking um, at that space, um, before the Hype or after ChatGPT came up? What are you, what has been your um, development in that area?

Speaker D: Yeah, it's great question. Obviously, what a time to be alive. Uh, in this space and particularly Jobs is so unstructured and unbounded and complex. Humans, um, can't often, uh, articulate what they want and agree on what good looks like. So unstructured text has always been a big challenge for us and so we have invested in what we thought were LLMs at the time. Um, so we did a lot of work two, three years ago around BERT models and fine tuning BERT models. So we have a jobberta model as we call it, which is fine tuned to our space. Um, so we've been working with what I would now call language models rather than large language models for quite a few years now, building things like models to predict how likely you are to be shortlisted based on your CV and profile in a particular job. Um, so we, investing in the space, we actually carved out quite a bit of budget to invest in fine tuning BERT models and similar models a few years ago. Um, so I think we were certainly in the space and understanding large language models before, you know, ChatGPT launched and all the hype really kicked off. Um, but then from there it's, everything's changed again.

Speaker B: So man, so much, so much and across, across everything. Um, so Jimmy, uh, I'll, I'll pass it to you in terms of how, how are you seeing the uh, the adoption of Geni from a market, from the market perspective in terms of both conversations, um, that you're having, what companies are looking for, uh, or thinking about discussing, uh, because I think that there's a fair bit of, of um, you know, brainstorming around this in terms of how to use it and then we can jump into the, the recruitment applications. But first, from the market perspective, how are you seeing uh, generative AI?

Speaker C: Uh, it's clearly on every, like everyone is talking about it at every event that I go to, every conversation I'm having with, with people, they're constantly talking about, you know, what businesses are using it, how, how quickly is it being adopted. Um, I think a lot of companies are sort of at the very, very entry point of being able to use the services, not too many of them actually have too many large language models that they're using. Obviously there's a few exceptions but um, you know, generally in the market everybody sort of wants to be touching on it. I'm speaking to people that aren't in tech, you know, accountants that are talking to me about I've been using, using it for this and that and everyone's sort of quite amazed with, with uh, just the basics of it really. Um, I think what's, what's quite interesting is, you know, speaking with engineers and data scientists is they're wondering how is it going to impact their jobs? You know, what, what can they be doing to, to make sure that they can sort of skill up for being able to go into that area? Um, what people are noticing is that it's, it's been really, really effective for a lot of the mundane tasks and, and sort of administrative work. Um, but that's kind of just really, really touching on the basics of it just right now. Um, but yeah, I think kind of what I'm hearing and speaking to people about is, you know, we're looking to bring in, whether it's a prompt engineer or whatever it is can we bring these people into our business. But maybe the education around uh, talking to people about how much actual experience that you can have that's going to apply to what is pretty new technology in the Australian sector is going to be really important. You can't be advertising. We're looking for someone that has 5 years experience with generative AI or anything like that because it hasn't really commercially been possible in the market just yet.

Speaker B: Exactly. And it's important m for organizations to be um, pragmatic or maybe realistic about how um, the, what they're looking for. Um, so. No mate, I completely agree. And um, how do you see the impact of Geni for recruitment during.

Speaker C: So it's interesting I speak to people and they're using generative AI for writing advertisements. The amount of people that are using it to put together their cv, um, to apply for X amount of jobs and sort of spamming the market with those kind of um, somewhat very uh, formulated cvs. But I think probably the importance in the people and the human element has never been as critical as it is now. Um, I think from a recruiter's perspective, moving away from transactional recruiting, understanding the people that we're meeting with the skills that they have that are going to be able to apply for particular jobs are Obviously on a piece of paper are really important. But you know, doing your due diligence, getting to understand, you know, various businesses that people have been in and how that's going to apply for whatever environments that looking for, that's not something um, that I've seen yet that, that can be figured out by um, Jen and

Speaker B: I. Yeah, definitely, Definitely not. And um, that's, that's one of the uh. One of uh. One of the reasons I really like uh, like as a hiring manager. Um, it's one of the reasons that I really like talent insights that you guys focus on the culture of the organization and focus on the culture fit of the candidate and how those are aligned also on the purpose of the organization and what is interesting for the candidate because I think um, that tying to the purpose of the organization, um, tying roles to their purpose gives a lot more longevity, higher uh, tenure and overall better satisfaction. And that's uh, a difficult matchmaking to do. Um, but yeah, one that you guys definitely do really well.

Speaker C: Yeah, thank you. Yeah, I guess sort of um, active listening and making sure that you're sort of paying attention to what are the needs of someone, whether it's now or X amount of years in the future is going to be really critical. But then obviously that applies to the right uh, kind of environments as, as you said that cultural, cultural fit is, is so, so important and it kind of, you know, typically guarantees a much longer tenure in a business as well.

Speaker B: Yeah. Grant, from your side, how are you guys thinking about the um, applications of generative gen AI from um, uh, sig's perspective?

Speaker D: Yeah, I mean I think a lot of what James has said, um, resonates with me. There's certain elements that can be automated or um, co piloted with this technology. And where that is possible we see that as our role. Um, but we still uh, very much agree that recruitment will remain a human centric business. So I think James and I talk a bit about that, the opportunities to work together on this type of stuff. Um, but from a SEQ perspective, uh, if I zoom out, we kind of think about it in three things. Is like the core information problems of an AI business that are like search recommendations, what job do I show you now in this channel? Um, based on what information I have. That's mostly what my team focuses on and there's a whole lot of applications there. There's also then that kind of UX experience side that you see and feel through the experience on the site. And then there's. Yeah. And as James mentioned the, the Back office, the input costs, the accounting finance, hr, ah, use cases that uh, organizations need to think about as well. Um, so on the first one we're thinking about well what is the customer problem, what's the job that needs to be done in recruitment and how can this technology help? Um, and there's all the obvious things around. I could help you write a CV better. Ah but as James mentions, you know that then reduces a whole lot of signal in the market. If you're taking a tiny bit of information and generating a heap of text, everything starts to look the same and our models would decay over time if we let that happen. So trying to think smarter about how do we increase the amount of information in the market, how does this allow you to express more intent in your own words and have us understand that how can we leverage things like company reviews to answer the first stage questions on culture fit, um, so that then we can give James the short list of candidates that he could really invest in and do the human touch. So from an information problem we think the technology unlocks a whole lot in terms of natural language understanding and understanding text, which is huge and our space. Um, I think on the experience side there's really interesting Things like writing CVs and other things that we need to think smart about as a centralized AI team, uh, working with other product and UX teams. We don't think we're really uh, we shouldn't be the bottleneck there. So we're thinking more about being a practice that can support teams to leverage off the shelf services for some of these things and really unlock and democratize AI uh across the organization which is a new. We always had that in our plan but now feels like the time. And then on the input costs back office piece we get a lot of requests from hr, finance, other groups. But my personal view at the moment is that will largely be well sold by vendors and the role we can play there is to support people to make the right decision, make sure they're using AI responsibly through our responsible AI team, um, and largely let the vendors play in that space.

Speaker A: Right.

Speaker B: That's great. That is great. So yeah, leveraging, um, yeah, I love so many parts of the answer, um, but yeah, can you tell us more about the internal product focus around um, creating AI as a platform or the product, um, to enable across the business, um, what are some of the either processes or decisions that go behind, uh, enabling that for the organization?

Speaker D: Yeah, I think partly it's a question of knowing where to go and who to Ask as to how do I start here. There's an element of setting up safe environments with the right infrastructure for our uh, product and engineering teams to operate in a way where they're not going to send data outside, um, seek, which we never do, but we want to make that as easy as possible to make the right call and access the infrastructure, um, choose the right algorithm, um, and get going. So there's a whole lot of just where do I start? How do I make it easy for a developer to solve this problem in a way that's safe and fit for purpose? Um, there's also quite important strategic and responsible AI guidance. So as I say, we could write a job ad, we could produce a product to use ChatGPT to write a job ad for someone, but that's probably not going to be good for the marketplace long term. So are we thinking about how do we add signal, not take it away, where do we take away friction and where do we add friction? So just helping with data literacy and understanding across the org, um, and that's both in sort of how do we think about data products and generating data for future use cases, um, and how do we use AI responsibly? So there's a big education piece and quite a lot of just admin support, uh, vendor or platform choice and infrastructure.

Speaker B: Yeah, that's excellent and thanks mate. And Jimmy, what do you think about that um, signal versus noise, um argument uh, in terms of with generative AI there can be so much more content or CVs, um, M. Um, job ads that can be created through that. Um, do you see it as that having uh, a uh, potential risk of watering down the information that uh, people can latch onto or um, do you see it as beneficial from a recruiting perspective? What are your views?

Speaker C: I think there's benefits and negatives. Um, but yeah, I think you're definitely right around sort of watering down the actual information. I think everything can kind of look the same. Uh, maybe a lot of the things that are really important kind of get missed, um, generally with those kind of job ads or cvs. Um, but you know, in terms of being able to structure things and be able to offer sort of an initial kind of almost like a template is really, really valuable. But it's, you know, it's the smaller or the finer details m that are going to be really, really critical and you know, from what I can see, often often get missed. Everything is quite sort of um, uh, very same same.

Speaker B: Right, yeah, because it seems like uh, what can come out of um, the models Is um, kind of maybe average or better. Slightly better than average, uh, across the board. But if somebody's an expert in an area they're going to be doing better. So in our case we would have technical expertise, but maybe not writing expertise, but blending the two. So it's not um, blending the two I think is a key. So it's not the only the outputs from the models that are being shared or submitted because then um, you are watering it down and having kind of like a, I don't know, like a B grade level. Um um, is what you're putting out to the world while you can do a little better with your knowledge. But I can understand that from a writing perspective you can get a huge boost by leveraging the technology. So yeah, I don't know. What do you guys think? I think there's kind of like working better together.

Speaker D: It's super interesting, right? It's a little bit like um. Because I look at, when I first look at this, I used to recruit for management consulting and the COVID letter was a really great signal. Right? Can someone write very concise cover letter that's kind of a solved problem now? And you look at that on first principles and say audits. I've a signal in the market that's gone. Um, but it's a little bit like the education debate. Um, if what you're testing for is now a solved problem with this technology, are you really testing for the right thing and is that a skill that's required anymore? So I think that's an interesting kind of behavioral acceptance piece that we'll have to go through and like in multiple sectors and some of it may be generational like we talked about the other day, self driving cars. And we look at a self driving car and think um, I would never let that take that control. Whereas our kids would probably look at it and think there's no way I would touch anything here, it could kill me. Um, so I think some of these behavioral trends, not just technology sharing trends will be really interesting. And the other one in this space I think is trust. So we're using Gen AI. Um, I think it's super powerful in summarization and um, there's also risks there. But when you're generating content, um, content, how do you build trust?

Speaker B: Yeah, yeah, because there's um, yeah, multiple, multiple issues there around you know, hallucinations, uh, reliability, trustworthiness, um, that um, in there's there's been a lot of discussions of different approaches and architectures on how to be able to limit that or improve uh, upon that. Um, and particularly as companies want to use um, Genai within their products using their internal data. Um, so some of the, some of the approaches that I've seen is people using uh, vector databases to consolidate their uh, embeddings of their documents and then they're feeding that through Genai when there's a related question and providing information only from the documents that is going in to Genai with the question, um, to kind of like try to limit um, the amount of information that Genai can draw upon for its answers. Um, but I think it's going to continue to be a challenge for a little bit longer. Um, but having better solutions in that and then having better infrastructure to keep the data, ah, the company's data secure and still being able to leverage models like that, then there's going to be kind of like the next wave of explosions where organizations are going to feel much more comfortable adopting the technology internally and for their customers. Um, and I don't know, I'm generally too much of an optimist and get excited too early but I definitely see that as a uh, as a bit of a game changer. Um, what do you guys think?

Speaker D: Yeah, I think um, you're in good company in the optimist phase. I think it's um, like. And the encouraging thing is that the improvements we're seeing around some of these risks as new models come out and we can talk about the challenge of keeping pace with um, just what's happening in the model space. But a lot of the improvements are actually around some of these risks which is really encouraging. And one of the challenges with vector recall which we think is really interesting, um, opportunity and it has a lot of architectural implications but you can always return something. Right. Whereas more traditional information retrieval um, tends to have a cap, um vector recall, you've got to choose a cutoff point or you could return anything. Um, and that can return if you have bad input, um, and you're always returning something that's a real risk. So something we're thinking about and what you described in terms of limiting the inputs that it can actually return as is one interesting solution there. Yeah, I share your optimism. I think these problems will get solved and I think like anything with new technology risks emerge um, and then you know, we use the technology to solve those risks and everyone's better off. So I think now's the time to keep pushing not to, not to dackle.

Speaker B: I love it. So true, so true. Um, and I uh, was thinking we can dive a ah, little Bit into the educational piece that's required around Genai and the, and the adoption. Um, maybe uh, I'll ask you Jimmy first in terms of what, what do you see people are um, how your discussions in the market, how are people thinking about it and approaching the um, educating the other business stakeholders or the enterprise around uh, gen AI users and risks. What uh, type of discussions do you, are you hearing from, from the market?

Speaker C: Um, I think is what Grant had mentioned earlier around sort of literacy and understanding the um, the risks and the benefits is, is super important. I think so many businesses are just kind of touching on it right now. Everyone wants to be implementing it. Um, not a heap of companies uh, necessarily got a lot of control over people using general generative AI across, you know, various, various teams. Um, so that's probably something that everyone's kind of talking about is what can we be doing to make sure that everything is going to sort of fit under the one sort of security um banner that we have and how can we make sure that everyone is kind of using uh, it appropriately? Um, and then you know, touching on things like bias and so forth as well. That's, it's definitely a hot topic but um, not everyone kind of has, has the answer just yet.

Speaker B: Yeah, exactly. Especially when it comes to the uh, yeah, general area. It's such a tough, such a tough question, such a tough um, problem space. Um, yeah. Grant, from your perspective, how are you guys thinking about the education behind the or to support the adoption?

Speaker D: Yeah, it's really interesting and it differs by audience. Right. So if you go back to peak ChatGPT hype in sort of, Jan said um, what I found and I don't know if this resonates with others listening but um, a lot of executive stakeholders and people around the business were exploring ChatGPT and got really excited about what it could produce. And a lot of the data scientists were like this is like this technology has been around for years. This um, is just normies catching up because now we've got, they've got an interface and they can understand it. Um, and I found myself doing that classic kind of leader, lean against the weight and the can do where you're like trying to educate the execs and business stakeholders that nothing's as good or as bad as it seems. And we need to think about this methodically while trying to kind of encourage our internal teams to really spin up some innovation groups and test this out because it did look really interesting. Um, and then on the risks challenge you've got this. What is a complex space to help people understand ethical risks and use cases and trade offs. And um, that was hard enough in a data science team. Now you've got to educate the whole organization. And I actually worry more about when you're not on the front foot with this. Um, yes, there's an opportunity you can cause some harm early on and you definitely got to keep an eye on that. But actually the bigger long term risk is people see the risks, they don't fully understand them and so they lock you down and you never unlock the impact of it. So if you can't get on the front foot as some of these risks start to emerge and some of the technology isn't paying off in the short term as much as people thought it would, if you don't have a clear strategy as to how you're going to deliver long term value and how you're going to manage risk, I think we'll start to see um, organizations kind of clamping down a bit in response to that risk. And so now's the time to get your investment case right and tell the story for the long term because we probably will see a bit of a dip in productivity and delivery once the quick wins are over. Um, and also how you're going to manage risk as some of this emerges.

Speaker B: Right, so true, so true. Yeah, I think um, it is the time to set up some good foundations so you can leverage it as an organization instead of having the, yeah, kind of like the walls come down and for it to go in the vault essentially. Um, Genii perspective. And um, I'll ask you, I'll ask you both and maybe I'll ask you Grant first, where, where do you looking into the future if we think about say five years, what, what would you like to see come out um, in your space as a result of, of generative AI?

Speaker D: Oh, uh, that's. It's such a, such a rough question for me. But um, in our space, if you talk about seek specifically, uh, I feel like jobs is the more you get into jobs, the more complex and hard and unbounded you realize it is. Which is what makes it such a fascinating space to work in, particularly in data. Um, but traditionally they've been like. Job search is still quite anchored on the role title uh, and experience as we talked about before, which in a fast changing world and even just for people's careers is not great. And it's also been pretty one way set and forget. So I write my cv, I give you a job search query and then I let that's all, that's all I tell you. Um, when James is dealing with someone he'll have a much more two way ongoing conversation. He'll probably float in some candidates to see what resonates. So I think getting um, from an online digital perspective, getting much more two way communication and rich context and trying to go beyond the role title to unlock skills and capabilities and culture to find the right fit, um, that's really exciting because that unlocks career paths that don't really exist today. And so a tangible example of that at seek, we hire a lot of product managers from management consulting backgrounds.

Speaker B: Right.

Speaker D: And we find they transition really well. That's not necessarily a common pathway in the industry.

Speaker B: Right.

Speaker D: But those things exist. It's not that these movements never happen, they're just not super common. And so if we could look into the tail and try and unlock some of those, um, there's huge value there and I think this technology allows us to do that. And the uh, benefits to people's careers and the benefits to the economy are ah, massive.

Speaker B: So true, so true. Um, yeah, I saw a stat the other day that um, the world GDP expect to increase 7% through the adoption of AI including Geni within that um, yeah, over the next decade I believe. So there's, there's definitely a lot of, a lot of upside from that. Um, Jimmy, from, from your side, how would you like to. What's the world you would like to see as a result of adoption of, of generative AI in your space?

Speaker C: Uh, the world that I would like to see from a recruitment perspective is as I mentioned earlier, sort of um, as it is now and kind of going forward making a lot of the mundane tasks quite sort of automated. But aside from that, um, I, I, I don't know how it's going to impact things around what I spoke about with um, people or the human element going forward. But being able to sort of assist in terms of um, you know, helping people learn, you know, create new jobs, that kind of stuff is, is going to be awesome from what's going to come from it. Um, you know, as Grant mentioned, you know, new career pathways, all sorts of opportunities that way. Um, but I still think what's, what's always going to be crucial is people to people connections and being able to uh, being able to learn more about each other, understand sort of what you're looking for going forward and what's going to apply to you know, various roles and new roles and new businesses that come out of it. Um, I think it'll be Interesting because everyone's so sort of bought into uh, the potential capabilities of it is I think there's going to be heaps and heaps of new business opportunities, new companies that come out of it and um, you know, the amount of innovation that's going to come from it's going to be really, really interesting. And um, you know, hopefully that can benefit recruiters, um, and everyone in the market.

Speaker B: Yeah, yeah, I agree. Yeah. I'm also really excited about how, how accessible gen makes kind of all of tech or a lot more of tech, um, that um, it kind of gives them the power for English to become a programming language in a way that like people can interact with it. And if they wanted to write, uh, some Python for example, like you can, you've got a, uh, translator from English to Python and you know, you can start to run things and um, and debug and et cetera. So I think it does, yeah, really great job in making more of technology more accessible, which is something that as organizations I think we need to do to be able to unlock um, more innovation and more value across the business that we're moving from a world where we had a few people with the skills in AI to having now, hopefully lots of people that have some skills, uh, and accessibility to that. Um, and yeah, I'm really, really excited by that. Um, I think it'll make a huge difference, as you guys have mentioned. Um, so guys, thank you so much. I think that's a brilliant, brilliant note to end on. Uh, this has been an awesome, awesome discussion. So I want to thank you both so much. Grant, Jimmy, thank you for your time. Thank you for sharing your experience, your perspectives, your knowledge with us and um, looking forward to the world, um, that you have described for us.

Speaker D: Thanks Felipe. Thanks Jimmy. It was a great conversation, really enjoyed it.

Speaker C: Thanks guys.

Speaker B: Thank you so much.

Speaker A: Thanks for watching this video all the way to the end.

Speaker B: I hope that you got a lot

Speaker A: out of this discussion. And if you're watching on YouTube, please like and subscribe to the channel. Um, so more people, people can find out about the challenges that leaders have

Speaker B: in the analytics and AI space. And that's what we're trying to share in better futurology.

Speaker A: Uh, so please like and subscribe. And if you enjoyed today's episode, uh, please tell your friends.

Speaker B: Thank you so much.