Paul Spain: Hey, folks. Greetings, and welcome along to the New Zealand Tech podcast. I'm your host, Paul Spain, back for our 1st live episode for 2024. Great to have 2 fantastic guests with us. Justin Flitter, from New Zealand.AI. Justin Flitter: Yeah. Nice to be here. Paul Spain: And David McDonald, from Being AI. David McDonald: Great to be back. Paul Spain: Great to have you both here. Justin, let's start with you as a bit of an intro where you fit into this this big world of of AI as we dive into an episode that we're really wanting to look at, you know, New Zealand's part to success in the world of AI in 2024. It's a Justin Flitter: Huge opportunity for us, isn't it? So, yeah, I founded New Zealand.AI in 2017, as a as a program, as a platform to Help business leaders and business professionals understand what artificial intelligence is, the different types of AI, and how businesses are leveraging it with in the workplace. And today, New Zealand.AI is where many Kiwi companies start their AI journey, and we provide a lot of workshops and training and strategy and consulting around, getting companies, integrated with different AI tools and building the skills that they need to create Sustainable careers for their people and and and competitive businesses, as well. So it's really exciting exciting time, and, really enjoy it. Paul Spain: Yeah. Great. And David? David McDonald: Yeah. So, I've been in the tech space for a couple of decades now, and in the, The, sort of AI and emerging technology space for about 8 years, and, had a had a couple of successful AI ventures in that space now, with with the likes of Futureverse, which has recently, had some pretty big wins in the market, and, and recently founded a new venture called, Bain Consultants and the Bain AI group of of companies, which will be, listing on the New Zealand Stock Exchange, in, March around March, through a reverse takeover. Yeah. Oh, they A lot of interesting things going on at the moment. Paul Spain: Yeah. Yeah. Well, I think it sort of really feels like we've got a An all star cast for this discussion today. Certainly from, you know, myself, AI has been, you know, an interest for for a long time. You know, I I guess this this last sort of, you know, 12 to 18 months, it's been a a big focus when I'm doing keynotes and and speaking to to varying audiences. But I I find where wherever I go that there's there's big gaps in in our knowledge. And when I look at the opportunities for New Zealand, it seems as though, there just There isn't maybe that sort of full understanding of how big the opportunity is and why it's really important that really every individual and every should be delving into AI at, you know, at whatever level is appropriate for them. And, you know, often that's, Hey. Paul Spain: Let's get our feet wet. Let's, you know, let's get started and get a feel for where the technology fits in for other others. It's It's that opportunity to build global organizations that are selling, you know, our expertise, our offerings to the world. And I guess if we can if we can work out how these things sort of fit together, and and maybe, you know, give some fill in some gaps for our audience today, I think that'd be fantastic. Justin Flitter: I reckon we can do that. Paul Spain: Okay. Let's go. So, if we if we sort of step step Back a little bit, there's a whole there's a whole lot sort of, you know, coming coming through, and there's a lot of technology that's obviously been released and made available in varying forms over, you know, over the last year. I would say most most of this audience, will be people that have been, you know, leveraging varying elements of, of generative AI technology. We've, you know, chatted a lot about chat g p t, which is, You know, has has taken, you know, a lot of attention. And then we've had the varying other sort of vendors sort of come in there, you know, Microsoft with what they've done to, to to license the, the technology, from OpenAI, Google, you know, IBM, and others that have been, you know, offering their their, things for a for a while in some forms and have been bringing new tools through. But if we look at at 2024 sort of specifically, I'm keen to kind of get a a little bit of a feel in terms of where we're gonna go this year, and and I guess in a sense of sort of making that, you know, accessible for for our listeners in terms of, you know, not going too deep technically in terms of where we're going, but, You know, what what the opportunities are that, that we that we land on this year. Maybe yeah. David McDonald: I I I think there's gonna be a big shift To, from, you know, the last year, we had a lot of sort of beta applications coming out, things that people can play around with and get a sense of What's happening and and how it all works. And then there's gonna be there's gonna be a shift this year towards more practical applications of AI. Gonna see it pop up in in in everyday scenarios. We're gonna see new user experiences come out that we haven't had exposure to before, we will be interacting with AI. And I think it's gonna be, there's a whole bunch of technologies that are gonna underpin that. So Multi modal AI, agents and and a whole bunch of other, fundamental, long term memory for for AI. Those sorts of technologies are are sort of being rapidly developed, and that's gonna unpin underpin a whole lot of user experience Changes for people, and they don't do lives. Paul Spain: Yep. And and just before we delve delve further on that, for listeners who are wondering, you know, what, you know, what do you what do you meaning by, you know, by those terms in terms of multimodal AI Yeah. And agents. Maybe you can just just break those down. David McDonald: Yeah. So multimodal is really exciting. It's, You've a lot of people have been exposed to things like chat gpt, which has been text. That's the mode that that, it operates in is text. And, but there's there's a whole bunch of other areas of research sort of sort of come to a head as well that has been beyond text like images And audio and different multimedia inputs and outputs. And once you stick all those things together, then you start getting really interesting combinations of things that you can do with AI. And then on top of that, with things like agents. So agents are at the moment, you have a conversation with chat g p t. David McDonald: Right? It's reactive, But agents will allow it to be preemptive, allow it to kind of process things on its own and understand the world around it, without your interaction, and that will open up a whole lot of other things. And then things like long term memory, but having the ability for the AI to Understand context of things that are not just within, the immediate conversation. That's gonna open up a whole lot of other doors and and user experience. Justin Flitter: Yeah. I think there's a convergence here of of different types of artificial intelligence. I mean, we've you you spoke about Kind of the text generation, then text image generation, text video generation with the things like, like, Peaca Labs is coming out. There's some astonishingly good generative AI content production going on at the moment. And we're we're starting to see New Zealand companies leveraging these tools to create Marketing and brand content. And soon, you'll see, you know, TVCs, maybe even full Fully fledged Hollywood movies coming out, which will be, you know, produced using generative AI, and that's going to upend, You know, industries. Paul Spain: Yeah. It's it's kind of mind blowing when you when you look at, you know, all of these all of these possibilities coming together. And, yeah, interesting. I was, you know, chatting as you as you do around these topics because it's it's so relevant, but I was I was chatting to, Yeah. A few people over over the holiday break around, you know, the latest iteration of the chat GPT app. And we've now landed in this world where, you know, as you're driving along, if you don't have somebody sitting next to you to you know, an assistant sitting next to you to start googling stuff and researching it and and giving you answers. You can actually, you know, have your phone and and and use your voice to, you know, to get those answers and to have that sort of conversation going back in. So So that's obviously a you know, that's a 2023 thing that's, you know, that's already, already here. Paul Spain: I know some of our listeners may not have come across that. So I I think we'll just give it a little demo. I'm not quite sure what we should what we should ask it. So has anyone got a either of you got a question that we should, fire at chat GPT. David McDonald: What about what do you think the future of AI user experience will be? Chat GPT: Imagine AI that understands your preferences and context better, making interactions more natural. Plus, We might see AI integrating seamlessly into daily life, like smart homes that predict your needs or virtual assistants that manage schedules more effectively. Also, there's a trend towards more immersive experiences. Virtual and augmented reality could bring AI interactions to a whole new level, Making them feel more real and engaging. And with advancements in language models, AI could become even better at understanding and generating human like text, Making conversations flow more smoothly. What do you think about these possibilities? David McDonald: That's a great self promoter. Paul Spain: Okay. We're gonna cut you off right we're gonna cut you off there, chat gpt. Thank you for, for being part of our part of our conversation today. Even just the quality of the voice there, the what it's able to Deliver. That's something that we've got access to right now. You know, if we if we look across the general populace, it's probably a reasonably small percentage that are having these These voice conversations, you you know, they're available to all of us. Justin Flitter: And it's it's that convergence of was that shift from typing And text. Right? We've got the in your interface for data input as a keyboard David McDonald: Mhmm. Justin Flitter: And it's now your your voice. Right? And and soon, or already, it's gonna increasingly become gesture created as well where you'll be able to, have gestured commands to to interact with, with devices as as well. And so, like, I've I've got a a a 6 month your old daughter, and I'm adamant she will never learn to type. David McDonald: Mhmm. That's interesting. Justin Flitter: I I just she probably will never learn need to drive, and she'll probably never need to learn to type. Handwriting? Yes. Make sure that she learns to handwrite. But, yeah, typing, I I can't imagine that's gonna be one that's, they're gonna be teaching in school anymore. Paul Spain: Well, that that's that's actually raised an interesting one because the few areas that we can delve into, education is is one of the one of the Key, here is I, you know, I believe not just the sort of the formal education sector, but, you know, All of us being, you know, lifelong learners. And and these tools, I think, are really, really helpful in terms of directing us and guiding us and and so on as as part of that. David, any any thoughts on, you know, where the opportunities lie? And let's Sort of focus this to, you know, towards New Zealand, where where you see those opportunities are, in education and both for, you know, our broader listeners, who who all need to be educating themselves and their family members and and so on, and sort of the broader business opportunities. David McDonald: Yeah. And, oh, well, to Justin's point earlier is is we are gonna see a fundamental disruption in the way that Children are going to be interacting with, with technology in the future. And you've just shown an excellent example of that. If I look at, you know, my, my my adopted son, he's 7 years old. When when we adopted him, he was, behind in school and was unable to, read or write, but he could could jump on Siri and Find any information in the world that he wanted. He had more access to information than I had when I was 10, 12 years old before the Internet happened. Right? And he's never learned how to to read or write, which is which is really interesting. So you look at the education system, now and what Sort of skills are teaching children. David McDonald: You've gotta look a little bit further into the future. We have so being being as part of being group, We, we're acquiring when we do the the listing in March. We're acquiring a school as part of that. So part of our remit is To invest in and acquire businesses that, in need of or will be subject to digital disruption through AI, which is pretty much every business. But education is a major focus for us and and our shareholders, because it's it is gonna be the thing that fuels New Zealand And and puts us, if we do it right, puts us ahead of the rest of the world, when it comes to AI skills and and the way that the world's gonna run-in the future. Justin Flitter: Well, I mean, I should Paul Spain: Is it I mean, is this practical that you wouldn't be learning to type? I mean, you're acquiring a school. So, you know, in that maybe you have some level of control, but government sort of expect, you know, certain things. We, you know, we might sort of believe, hey. We can, You know, do x y zed, but, yeah, governments, of course, can be mandating that you that you need to do sort of something different, and they're often, you know, quite, you know, behind shall you know, shall we say. David McDonald: We've had interesting conversations, with with government, across our group. And, you know, there's there's new things kind of like partnership schools coming out, which are gonna enable us to explore in better ways to do things. You look at the school that that we've, that's coming into our group. They've they've looked from the beginning at different ways of doing it. So that's they've a history of that now. Paul Spain: So it's a private New Zealand high school, is it? David McDonald: It would so so they from primary to high. Okay. High school. Yeah. And and being able to take that model from from that school and and and project it out into, into this new world that we're going into, where they're already thinking about, what skills are children gonna need to succeed rather than just just teaching, you know, you've gotta learn how to write these things in cursive and all that sort of stuff that might not necessarily be skills for the future. As that's that's gonna be important for New Zealand schools to do. I think things like partnership schools are gonna, help with that a lot, and Talking and and and, opening up those conversations with the government and things that, like, Justin's doing, going out there and educating New Zealand businesses And and government on on what this stuff means. That's really gonna be important. Paul Spain: You know, Justin Flitter: and I think you've we we wanna be teaching the fundamentals, so how things work And then sparking curiosity. And no matter what your passion, what career or interest you have, If you've got the fundamental understanding of what artificial intelligence is and how the different elements of it can work, Then you can see opportunities to leverage AI in whichever skill set or professional or personal field that you want want to be in. And that's really important. Understanding what data can we be collecting. What are the interactions or the tasks that are repetitive that we can automate or augment? And so the pace of life the pace of not necessarily the pace of life. The pace of doing stuff is going to change. And and that's why I kinda think about, you know, your typing speed is how many words a minute, Paul? I I'm pretty sure that we can all speak And generate, content a hell of a lot faster using our voice than we can, typing. And I just think, you know, your your point earlier around, the transformation of of UX. Justin Flitter: You know, we're we're gonna see devices that are going to completely change in their shape and style, and and wearables and the way that they project or augment other elements, around us. And I think they're largely gonna be, gesturing and and and voice enabled. The way that we communicate with each other Using voice and gestures is exactly the same way as we'll be communicating with with computers, and that's, you know, what we've seen With ChatGPT over the last 18 months, that's made it so accessible to the general public is we're interacting with these machines In a very human like way, and that's the that's amazing and scary at the same time. David McDonald: What's kind of the goal of of user experience, right, is to make it more natural and human. Mhmm. We've seen the evolution of that from this mouse thing that, you know, Was kind of kind of a good way to interact with computers all the way up to, you know, Ipads and iPhones, which was much more, you know, finger gestures, Much more user friendly and kids can jump on and learn it pretty quickly to a point where it's just completely intuitive. Mhmm. And that's been the trend of user experience. But And, you know, it's not not too long before apps won't exist. That's that's another thing that people, an area of disruption that, You know, people don't necessarily see coming is that, you know, if if, if AI can write code in real time for anything you describe and ask it to do, Why would you want to build an app? Right? You can it can write your the the application that you need to do in that moment and then just get rid of it. I think those are places where it's gonna disrupt, existing. David McDonald: What we thought a couple of years ago is high-tech businesses are gonna be completely disrupted by AI in those sorts of ways. Paul Spain: Yeah. I think that's an example that would probably, you know, completely blow Low low people's minds. And as I start sort of, you know, delving into it inside my head, it's like my head's exploding into small pieces, all over the room, in my mind anyway. Yeah. The I mean, there there's so much of you sort of, you know, break break that down in terms of Yeah. I guess, the reality of the you know, we do like the consistency of being able to tap in and and use our apps and use, You know, use different things. But I I get I get that it that it's changing. I'm not I'm not sure how quickly on the fly we're gonna be rewriting this, you know, how we operate. Paul Spain: But we definitely are, and and I guess how how it evolves is kinda Open to as open to a lot of imagination and and interpretation in terms of, you know, how it ultimately plays out, but it's obviously Probably bigger than what any 1 individual can put together because for sure. We only kinda you know, we'll only see it from our own perspectives. Right? Justin Flitter: So far, For the most part, in the last 12 months or so, we've seen lots of different companies hammering some AI into their product. Right? They're just doing the integration, we're putting some generative AI or some kind of functionality into their software platform. And where we're gonna start seeing the real Transformation is when new ideas are developed using generative AI from the ground up. Right? We we wash away all of this, The last 20 years of design structure and thinking and the way that we build and the way that we code And we start doing it using entirely generative AI processes and integrations. And and that's where a lot of these, traditional legacy technology platforms should feel pretty concerned because it's Quite likely that a competitor could be stood up in a few months, let alone the fact that they've been building their legacy platform for, you know, for 10 years. So it's the speed of innovation is is going to continue to accelerate, but at a much more astonishing rate than, Yeah. David McDonald: And that's why that's why things like, New Zealand.AI and Being AI need to exist is because There are a whole lot of leg legacy businesses out there, a whole lot of companies out there that don't necessarily understand what that disruption is gonna be for them. Paul Spain: Yeah. David McDonald: And they need to do something about it. They need to do it pretty damn quickly. Paul Spain: Yeah. And I and I look. I think it it's gonna vary a lot according to the the the types of entities and so on. And there's still a lot to be said for getting our feet wet and using the technology. I drove through Burger King, I don't know, 3, 4 days ago. And as the 1st time I've driven driven up and not spoken to a person. And so in New Zealand, they've got Now 3, 3 stores where you drive up and it asks for your order, and you're recognizing hold on. That's A very computerized voice, I believe. Paul Spain: Right. And it's basically operating exactly the same way as a normal as a normal person, asking you the question. You know, what do you want? Want a couple of ice creams? It's a hot day. And and it did a pretty good job. Like, it wasn't it wasn't completely tweaked. Like, at the end, it said, that'll be 2. I'm like, what? 2ยข? $2? You've left a word out there. But, but something that probably rolled out reason reasonably quickly, and it's, you know, it's probably helping them deliver, maybe with, you know, lack of workforce, whatever, you know, whatever the the things are. Paul Spain: On the flip side, They're possibly dropping the user experience from an individual's perspective who's used to dealing with a person. So these are some of the things. But what I like is that They're trying out the technology. They're working on where it fits. Yes. They're not completely disrupting their business, but I think, You know, most of us in the technology world can imagine, you know, a Burger King that you could drive through that has nobody working in there or a Burger King you don't have to drive to at all because the the nearest one drops it to you via drone. There's all these sort of, you know, possibilities. Right? Justin Flitter: So there's a keyword disrupting your business, And this is where a lot of companies probably stall with with artificial intelligence as they think that it's gonna have to change everything that they do. Right? And it It it doesn't. Most of the applications that we're seeing in businesses around New Zealand, certainly the work I'm doing, Small tasks or small workflows, and we are finding hundreds of use cases. When we deepen dig into the different teams and how they do business and the way that they operate, we can identify hundreds of different use cases And then start identifying what's the solution that we might put into to solve some of those or the different integrations and workflows. But it's not about completely reimagining the way that you do business. It's about lots of incremental improvements, which, You know, together is over time is gonna create new superpowers for the for your business, but it's not necessarily a com you know, a whole of company, transformation where you're ripping everything out and starting again. It's it's just it's a it's it's it's it's Fast, but it's gradual. Paul Spain: Yeah. There's I guess there's opportunity for both, isn't there? But when we when we look at it, there's not a lot of Airbnbs in the world. Right? There's there's not a lot of, Netflix's and and and and so on. So there there will be some that will have those, you know, completely mind blowing ideas that completely destroy or, you know, reshape a particular sector or become, you know, new areas. But The reality is for for most people that are listening to this, you know, the immediate opportunity is is, You know, how can we apply some, you know, some aspect of artificial intelligence to to provide those sort of smaller incremental things? And if you keep stacking those things up on a, you know, on an ongoing basis, actually, before you know it, you do end up with something that is quite different to where you were a year or 2 ago. Justin Flitter: How can we use AI to become more competitive, to increase our productivity or efficiency, to bridge the skills gap because we know most companies are struggling to find the right people when they need them. And maybe it some of it, it might be reduced costs. So there might be some roles, and this is that kind of what I've mentioned earlier around sustainable careers is that The roles are going to change. There are going to be aspects of your job which are going to become automated or augmented with artificial intelligence. That doesn't mean that your entire job is nonexistent. It just means that your role is going to is going to change. And and so that might free you up to do Higher value work or more meaningful work that's gonna contribute to the company in a completely different way. And the task that you have to do every single day that drive you batty because they're just repetitive chores. Justin Flitter: You can kiss them goodbye because the machines will be able to take Over those for you. Paul Spain: Especially if you don't like doing them. Right? It's a bit of a different thing if it's like, hey. I love writing, and I'm really good at this, and it's It's, you know, it's my passion. Justin Flitter: That's your classic that's your craft. Yes. Paul Spain: So there's that balance. I don't know if Jo Ellen's mic is on, But Jo Ellen uses AI tools and has done for, you know, for some time within the studio. If you're able to give us a, you know, a quick summary of of how It helps you and how that's kinda impacted your work. Jo-Ellen (Producer): Hi, Paul. Hi, team. Yeah. I found that AI has really been helpful for me Just to, as Justin said, get rid of some of those menial tasks that are very repetitive just by using a handful of tools from a podcast point of view. I can send it the data. It goes through. It gives me a whole bunch of options, and then I can media and curate The content that I want to, put out there. So I'm still having that touch. Jo-Ellen (Producer): I'm still being, as as Paul said, being able to do some of the creativity, But it just takes a lot of the leg work out for me. Paul Spain: Having a tool that, you know, can basically listen through the whole episode and then kind of summarize. Like, that's actually a really time consuming thing to to write these things up and so on. And at the end of the episode, my mind is is full, and I'm not kind of in a state where I've kind of feel like, I can write up the summary, you know, of that. We've got editing and stuff to get on with. So that idea to be able to hand it over and it to carry some of that workload is is just So helpful. Justin Flitter: It's not role replacement. Right? It's it's augmentation, and there's still a human in the loop here. Right? That's it's still really important. It's not like we're just Letting the AI go off and and do its thing. We're providing some tasks. It's generating some outputs. We're checking it. We're doing our human stuff on Layers on top of it. Justin Flitter: And so it's it's been that digital assistant, that's supporting us at an individual Level, at a team level, or at a company level? David McDonald: The trick for New Zealand businesses is is gonna be navigating the sea of Tools Yeah. That are out there. The sea of innovation. It's a daily barrage almost of new new technology being developed, new Studies being put out, new white papers to read, new products coming to market. What ones are the the genuine ones? What ones are just A lot of fluff or hype. What ones are gonna apply to your business? What ones are gonna disrupt your business? It's a it's a massive market. If you wanna, you Now condense the 1st 15 years of the Internet into 12 months. That's what I think we're gonna see that Yeah. David McDonald: That kind of scale. Paul Spain: And and you you raised some really good points here. You know, I think around, you know, the the the role that my business plays working with helping small to medium organizations to leverage technology well. And one of the most common things we see is that desire to adopt some new technology. Oh, we found this really cool app or this platform or we need 1 and how do, you know, how do we pick them? And it's always, it's always quite a, you know, quite an interesting journey, through that. And sometimes as an organization will say, oh, no. We've got it. We can We can figure this out ourselves, and then, you know, we're looking at, what about your data, you know, privacy? What are they gonna do with your data? If you're handing over All your stuff to some random firm that's opened up and, you know, on the other other side of the world, and suddenly they've got all your data to to do the smart stuff with. Do, you know, do you trust them with what they're gonna do with the data that you're feeding into it? And then, you know, How long is that firm gonna be around? Are you gonna really get reliant on them? And, actually, there are, you know, they're a tech startup that only had a sorta, you know, small runway in terms of, you know, the finances available to them. Paul Spain: That's run out. You've got all your stuff in there, and, they're gone. Or, you know, there there's just there's so many things that we kinda see, you know, with with the stuff. And, actually, we need to be able to, you know, Move really quickly, but also be making, you know, good educated, you know, choices with with what we adopt. Right? Justin Flitter: Fancy text all fair and well, but you've gotta understand what's the business problem we're trying to solve with us or what's the new interaction opportunity we're trying to create And then go and find you know? It's so it's it's all or on the other hand, it is foster your curiosity And just play around with whatever you can get your hands on and figure out the ways that you can use that to short shift what you do at work. So So there's kind of 2 different, 2 different ways that companies come to come to me with with with questions on on their journey. And, You know, we we help them navigate through that plethora of of of new tools. And the exciting thing is is that there's lots of, New Zealand made, New Zealand designed and built AI for business tools that are that are coming on the market. And, you know, that that means we've It's it's gonna become more accessible from a cost and support as well as, you know, just being able to to to buy and use KiwiTech rather than having to continually look offshore for for a solution. So Paul Spain: Yeah. And we definitely wanna be you know, we wanna be encouraging that. Right? We want, you know, Kiwis to be building things for Kiwis, but, you know, even more so for the global audience. Yes. If they can build, you know, tools that are gonna help, You know, all organizations or, you know, broad range of local organizations do better, but we also wanna be we wanna be exporting and kinda leading the way. So what are your thoughts, David, on how we how we how we do that? How can how can New Zealand sort of, you know, really lead from from that perspective? David McDonald: We've got great examples in New Zealand of of New Zealand New Zealand is doing that. You know, we've got you know, look at look at the stuff that 0 has done. Look at the stuff that, you know, Peter Jackson and and, those guys have done, and and we've got view. Yeah. We've got we've got fantastic and actually even Futureverse, you know, $1,000,000,000 New Zealand company out there leading the world in in, several places and and including MusicAI, which is really, really interesting space. And it it takes it takes a global focus. It's like you take you take that New Zealand ingenuity, that that thing that we've grown up with, you know, the thing that I'm proud to be a Kiwi 4. Kiwi ingenuity. David McDonald: Kiwi ingenuity is, you take that and you direct that externally. Right? You you don't look at what What, what, you know, you can do just for the guy down the road, but you look at what you can do for the world. And we we live in a world where that's possible now. You know, the borders are are so much thinner than they used to be. We're not an island anymore in a lot of sense. You know? We're we're connected to the rest of the world, and that's that's what The Internet's enabled us to do and and send modern technology. And, you know, you gotta think About what your your idea can do for the world, outside of New Zealand and and just go for it. And I wanna support, New Zealand startups part of Part of our, within, Being AI is, we're gonna ventures, specifically to do that, to look at Ventures that we can we can support. David McDonald: I wanna see Kiwi companies, having that global focus and And taking our our powerful engine ingenuity and and putting it out there. Paul Spain: Yeah. And it and it does seem as though there is that, growing Availability of, of funding to, you know, support startups and and companies in this space. And I think that that is gonna be really important because we're gonna need to move quite quickly. And, yeah, opportunities, You know, potentially gonna be, gonna be lost if if firms can't, you know, tap into the appropriate sort of, you know, capital And and that, you know, broader expertise in the in the varying areas from getting out and sort of selling to the world to, you know, just having the Right? Sort of leadership skills, which caused New Zealand's so much, you know, stronger and then where we were a decade ago. David McDonald: Oh, for sure. And have an ecosystem approach, you know, Kiwi companies helping Kiwi companies. Mhmm. If if we just if we just turn inwards, then Wind up, you know, fighting over a few scraps. But if we we look to to the global, market and we see what all of the New Zealand companies can do together, a thousand is a a quote from my brother and, that he said a few years ago, Aaron McDonald. A 1000 Zebras is better than 1 unicorn. Right? So we had, you know, a 1000 realistic projects all going in one direction is better than a mythical creature that doesn't exist. Justin Flitter: We we're so well positioned, though, aren't we? I mean, we've had, like, the last 20 years of, you know, great New Zealand technology been designed and built, Bordered around the world, a lot of wealth being transferred back to employees as through ESOP programs and investors. And so that whole VC market in New Zealand is, is is is well established. We've got, you know, Incubators and universities and research. There's a lot there as well as all of the international, support networks which are there to, help companies land and expand. And now with artificial intelligence, it's it's the The the tricky part of this is coming up again. Well, like, with any startup is coming up with the right idea. Right? And being able to have that intuition, and The ballsiness to go with that intuition, with that, all that support. David McDonald: Ideas are usually in my experience, the easy part is executing. It is, pulling together of the resources Mhmm. And sticking with it until the job's done. That's The hard path of getting something off the ground. Paul Spain: I think it probably depends where you sit because within a, you know, within, say, a small organization like mine, Yeah. Yeah. It's it's there's that combination of finding what are the right ideas that are gonna add that immediate value. Some of which can be quite quick, and you can have an idea in the morning and put it in place in the afternoon, right, in a in a smaller organization. But I think of some of the The bigger things that that we're trying to do and, you know, the ideas are there. And it's exactly as you say of like, we don't quite have enough, you know, resource on the developer side and and and so on to have been able to, you know, execute, some of the things that are, you know, that are in the in the plants. And so that you know, that's part of what we'll we'll work through and and, you know, hopefully execute, you know, over the over the next sort of 3, you know, 3 to 6 months. But some of these things, you know, the quicker you can get going and and you're that that execution ability, can make a difference between whether you're in business and and not in business. Paul Spain: Yeah. Definitely. Yep. Justin Flitter: Trying to design and build A new AI powered something must be daunting given the fact that OpenAI is just Eating up all of these different ideas that, you know, the these big tech companies are just looking at all of these Little use cases and things that are popping up all over the tie all over the place, and can quickly incorporate that functionality And to what they're doing. So, you know, being able to have that creativity and foresight to kind of really see, a a a a use case or a scenario or a new business idea that's not likely to get washed up into Some other like, that must be, pretty pretty challenging. Paul Spain: Yeah. I I don't know. I think there's, you know, there are these sort of smaller niche New niche tools and some that are becoming available. Lots, you know, lots and lots of them as we sort of talked about before. You know, that won't necessarily be available from a big player that, you know, we probably all can tap into in different you know, in different ways. And, you know, I I think of, you know, what we've done in in the, you know, in the podcast work with, Podcast NZ. And, you know, we found a toolset that is as saving a whole lot of time. Yes. Paul Spain: You could go out and do it, you know, 3 or 4 or 10 or a 100 other ways, but, someone's built something that's quite, you know, niche and fits in quite tightly with our, with our workflow. And so, Yeah. I think each organization will have those unique things where where that can, you know, that can make sense. And then you've got, Obviously, the the big players, and I think we do wanna delve into sort of the role of, you know, OpenAI and Microsoft and Google, the open source things that are that are coming through as well. But I I was chatting to, chatting to, somewhat, I guess, was an off the record, bit of bit of commentary, but around the the role of of Microsoft's, you know, Copilot within 365, which isn't broadly available yet. But the the the sort of summary, which I'm I'm sure they'd be okay with me sharing, was was that, their sort of strongest use case of this, you know, big toolset that's become available, to some organizations within the three six five world. Their their biggest sort of value they were getting out of it, was its ability to summarize their meetings and who needs to do what at the at the end of it. But, you know, these things are all kinda gonna Gonna vary in terms of what we need. Paul Spain: You might find, well, it's not worth paying for a big toolset just to do that one thing because we might have another tool, or you might have something that ties in more to your workflow. So, you know, I think we we will probably find a range of of things to tap into. Justin Flitter: I think every company, know what size has a problem with organizational knowledge. Sure, you can save documents in SharePoint Or Google Drive or wherever you wanna pull it put put put it all on your local machines and, you know, good luck finding anything. Now with generative AI and some of these tools like Microsoft Copilot or, or Smart Space, You know, we've got the ability as generative search and generative AI to not just find documents, but find ideas within documents And then create the workflows off the back of all of that to create tasks or to create new documents using generative AI. And to be able to then, record, store, and use Organizational knowledge in a much more seamless way that we've not been able to do elegantly Until now as as well as we can now. And a lot of, you know, companies, people are using ChatGPT Four or three and a half? They're using their personal accounts to do company work, to ask it questions, to come up with Structures and documents and ideas for the work that they're doing, and all of that knowledge is all going out of your tendencies. It's leaving the business. So we need to be bringing that GPT functionality, that generative AI functionality As enterprise grade tech within your tenancy so that we can utilize all of those those ideas and build on them so that when someone Resigns or leaves the business, that a lot of that stuff walks out the door with them, and we need to better, you know, retain that as as possible, I think. David McDonald: I think it's as interesting to even look at how we're using AI when you when you talk about that, Process flow that you just went through. You have somebody using AI to create a big document, giving it a summary, sending it to someone else Who's using AI to reduce that document into a summary? Right? And that's what's that's going on. It's surprising at the moment. Chinese whispers. Yeah. A little it's yeah. That and and you get a little bit of that effect as well. But it is funny that we're using all this This power and technology to blow something up and then shrink it down again, on the other side, when when in reality, we we might need to look at ways that We have AI communicating to AI. David McDonald: So I have an AI on this side that communicates to, an AI on that side, and it Summarizes the information between the 2 of them that we need. So those sorts of, advancements are are gonna be the ones that are gonna change the way that we do things. I think we'll see a little bit less of this, you know, double working that's going on at the moment, in the future when when those sorts of innovation It starts. Paul Spain: Yeah. I think in the same way, we've got you know, we've had apps that talk to other apps through through APIs and so on. You know, we will we will get these more, you know, more seamless, you know, communications and and, and integration. Now, before we wrap up, I think it it would be good just to sort of, I guess, you know, we don't have a lot of time, but sorta lightly sort of talk about the role of of the different, you know, players. And I think we're probably gonna need to come back into another Another session to delve into some of the other, other aspects. Obviously, we, you know, we had OpenAI kinda, you know, just about imploding and then, and then, you know, kinda coming coming back together. Justin Flitter: Wasn't that some theater? Paul Spain: In in in recent weeks. And, and so, you know, so that's happened sort of, you know, the end of 2023. Things didn't now back together. Google, Yeah. Really, really interesting to to see some of the or to see glimpses of some of the work that, that they've done and, you know, their, Confidence, arrogance, however you wanna sort of put it in terms of how they're they're better than, OpenAI with, with what they're, you know, they're delivering now. Them sort of talking about, hey. What we've built is multi multimodal versus what, OpenAI, doing isn't and, you know, there's probably a whole lot of debate around whether that stuff's, you know, true true or not. But, you know, how how do you both sort of see for for those that are listening and wondering, well, you know, where where should we be sort of placing bets or who should we be looking to in terms of, you know, whose technology to be, you know, leveraging this year. Paul Spain: How do you, you know, how do you make those sort of decisions? David McDonald: So, as far as I I think I think we got a lot of water to go under the bridge before we start deciding on on who's gonna be a winner, in this space. One thing I know for sure, so if I was to be placing bets, so it'd be on any anything open source. Open source, community is gonna outpace proprietary community all day long in this space. The pace of innovation that the open source community can, can do the the different ideas and the different hypotheses that they can test on a daily basis is greater than 1 single organization can do. So that my my bets on, on, open source communities. You look at what Google's doing. They're embracing they've they've kind of got There there are an internal, proprietary stuff. They've got career contributing towards open source stuff, and they're embracing other open source Products in their in their cloud suites. David McDonald: I think that's a great approach. And you see Amazon's also, having a a similar approach, and the way that they're looking at. So I think that's gonna be the way the way it's gonna pan out. The these most of these organizations are hedging their bets, just like we should be. You know? Paul Spain: The competition's good, though, isn't it? It is it is definitely good. In a great position to to We're not sitting here and saying there's just 1 dominant player, and they're going to be able to kinda extract Unlimited amounts of money out of us. We actually we've got a a, you know, I think a really good level of competition. Although Yep. You know, there's obviously some clear, you know, leadership the moment. Justin Flitter: And you've got some some new new companies out there like Hugging Face and, you know, there's all sorts of new organizations, that are that are stepping up with with models and functionality, which, again, will is is great for competition and and great for innovation. David McDonald: And Hanging Face joining forces with AWS. I think that's a powerful Mhmm. Combination there of of capability that Justin Flitter: we can do. Let's not Forget out of the conversation. True. Either with with with some of the the models that that they're developing, I think certainly one to watch. And again, it's you know, they're they're they're gonna have different strengths and weaknesses in terms of, the the skills that they're gonna be really good at or or tuned for. And I think I think might have mentioned earlier around, Look. I can run a a a a a 7 and a half 1000000000 parameter large language model on my phone now. Right? And so we can then start training, and we're getting we're getting language models that are getting smaller and smaller and smaller. Justin Flitter: Right? So it's 2,000,000,000 parameters and being able to fine tune and train those. And I think, you know, individuals and companies will now start looking at at at how do we How do we leverage some of these these smaller open source models, and and train them for specific tasks that we want to do? And so it's, Again, it's getting into that menu way of things that we wanna be able to do. It's not about this massive transformation program. It's about thinking about What's the problem we can solve or the new thing that we can create, with some of the data or, you know, things that we can collect. So, yeah, it's it's It's a it's a great time when to be alive when you can, you know, be creative with some of these new and emerging techs, and they're not That difficult to start playing around with? Paul Spain: Yeah. Well, as we as we finish up, I'd like, you know, to say thank you to the both for for joining the conversation. I do think there's there's more we're gonna have to kinda keep coming back on on this conversation. We're great to involve you, both during the year. And, of course, there's there's been a lot of discussion this last year as well. So, you know, there are there are areas that we haven't sort of delved into in terms of, You know, the role of of government and legislation and, data privacy and, you know, the robots taking all of our jobs and and these sorts of discussions. But, you know, we'll keep coming back on on these things. And and also looking at how do we solve some of those bigger problems that, you know, we have as a as a country. Paul Spain: And, you know, what's what's the role of AI gonna gonna play there? But, But, yeah, thank you both for joining the show. A big thank you to our show partners who who make all of this possible. 1 NZ, 2 Degrees, Spark, HP, and Gorilla Technology. And look, you know, the the nature of AI means everybody's sort of involved in one way or another. So if you work with those companies or, you know, those companies are are worth looking up in terms of what they're doing from an AI perspective. I see from, You know, HP, they're starting to sort of build in an element. You're buying a machine, and we'll see more of this, a device that sort of comes with, you know, AI chips or AI sort of, you know, software element to it. You know, the the, the Telcos are no longer you're just telcos, of course. Paul Spain: They're they're offering, you know, a range of things. So, there are plenty of sort of touch points with organizations that can help us, including your own companies and, and and mine. And, of course, I guess the 3 of us are Are, out speaking to audiences as well on this subject. Absolutely. You know, I guess the encouragement is for listeners as to you know, You probably everybody will have will have made some steps forward in 2023. But we need to kinda go next level with that for for 2024, and not kinda kick back and go, oh, okay. Yep. 2023, that was the AI year. Paul Spain: 2024, we've we've gotta, you know, work on work on other things. I think it's, know, it's gonna be a key part, of our future. Justin Flitter: Agreed. Absolutely. 2024 is the AI becomes general. Paul Spain: Nice. Now where do where do people track you down, Justin, if they're, wanting to get in touch? Justin Flitter: Oh, that you can visit New Zealand.AI or, connect with me on LinkedIn, Justin or or, ex, Justin Flitter and and New Zealand.AI. I'm everywhere. Paul Spain: Easy. Easy. Yep. And David? David McDonald: Likewise. So, Being consultants, and, connect with me on on Twitter as well, David McDonald, and on, sorry. There's a X now. And, and LinkedIn, obviously, David McDonald. Paul Spain: Yeah. That's great. And for folks wanting who are listening to the audio, make sure you tap in to get access to the live feeds that we we do most weeks, If we're all able to get together in the in the studio, so the best way is probably following myself, Paul Spain on LinkedIn. But we do also stream across on x, Facebook and YouTube. And, of course, you're, watching the live stream, then make sure you fire up Spotify, whatever you use for podcasts to, subscribe, to the weekly NZTech podcast, audio. Alright. Thanks, everyone. And, thanks again to our show partners, and we'll look forward to catching you again next week.