Engineering the Future of Productivity

• Australia is showing how a rich country gets poorer
• UNSW Sydney lands $3m Defence project to build next generation AI
• ‘Dangerously unprepared’: UNSW expert says Australia must regulate AI and invest in research
• Humanoid home robots are on the market – but do we really want them?

Neil Martin  00:06

Welcome to UNSW Engineering the Future podcast. In this episode, we'll answer a big question: How can Australia best embrace artificial intelligence to not only lift productivity, but also raise living standards and protect jobs?

Sue Keay  00:23

It's really important that Australia looks at how we can harness artificial intelligence to help lift productivity, or else we do risk stagnating or even dropping in our productivity, which will directly impact on our living standards. It really isn't an option for us to just bury our heads in the sand and hope that this technological change won't impact on us. We have to figure out how we can harness the benefits.

Neil Martin  00:48

That's Dr Sue Kay, director of the UNSW AI Institute, who says, as a nation, we must be focused on unlocking the true potential of technological advancements such as artificial intelligence and advanced robotics. On engineering the Future, we speak to experts and researchers who are embracing cutting edge ideas and pushing the boundaries of what is truly possible. Join us as we discover how world-changing action starts with fearless thinking on Engineering the Future of Productivity.  

Hello and welcome to Engineering the Future of productivity. My name is Neil Martin, and I'm a journalist and STEMM communicator working in the Faculty of Engineering at UNSW. Joining me today to discuss how artificial intelligence, both physical and non physical, could be vital in boosting Australian productivity is Dr Sue Keay, director of the UNSW AI Institute. Sue is also a founder of the Robotics Australia group, and her expertise lies in leveraging robotics, artificial intelligence and automation, to solve complex challenges across various sectors. She has been the robotics technology lead at Australian copper mining company OZ Minerals, and is an industry Fellow of the Australian Academy of Technological Sciences and Engineering. Many thanks for being here, Sue.

Sue Keay  02:17

Yeah. Thanks, Neil.

Neil Martin  02:19 

So I might just jump straight into this because I saw a 2025 report, I think, from Deloitte as part of my research, which said that since early 2022 Australia's labor productivity has fallen by 5.7% that sounds quite significant and also potentially worrying. But can you explain why productivity really matters, both to the country as a whole, but also for everyday Australians who might be listening to this podcast?

Sue Keay  02:51

Yeah, well, productivity can be directly linked to our standards of living, and so typically, if you start to see productivity stagnating, then that can influence how much we can afford to buy from other countries, and generally, how comfortable we feel with the wage that you know we're all receiving. Now, the problem with productivity is that there's a limit to how much you can improve a country's productivity just through labor productivity alone, because I'm sure people probably don't feel that they can work that much harder than they currently are. And so the main way that we've seen a lot of productivity improvements in the past has been related to technological advances. So if you think of electrification or when desktop computers became widely available, that is one of the main mechanisms that you can improve a country's productivity, which is why it's really important that Australia looks at how we can harness artificial intelligence to help lift productivity, or else we do risk stagnating or even dropping in our productivity, which will directly impact on our living standards.

Neil Martin  04:04 

And can you go into a little bit more detail about how that increased productivity improves living standards for the general population? People might think, oh, that's all to do with companies and how much money they're making and what profits they have. How does that benefit me?

Sue Keay  04:20

Yes well, it's not just companies. It also improves the efficiency of government services, for example. So one of the biggest costs to the Australian economy is the amount of money that we invest in healthcare. So if we can do that more efficiently, that means that there's more money in the government's coffers that then flows directly back to the public. But your point about companies benefiting from productivity is certainly true, and some of those benefits may be passed on to employees, and some of them may not. But in general, if Australia is in a strong position and able to export to the rest of the world at competitive rates, then that tends to keep jobs on shore, which is good for everybody. And then in terms of how the profits of that improved productivity are distributed through society, that's actually a, you know, a question for our government and unions and, you know, collectively working with industry to make sure that there is hopefully less income inequality. Although what we have seen is that across the world, there does seem to be increasing levels of income inequality, and we should all be concerned about that, and what mechanisms we can put in place to stop that from happening. That has been happening regardless of this artificial intelligence transformation, but unfortunately, that technology also has high potential to accentuate those changes, with a lot of the benefits potentially accruing to a very small proportion of people. And that's something we have to be very sensitive of.

Neil Martin  05:52 

You talk there about the technologies, and you mentioned it before that that's the opportunity, over and above increased labor productivity. Can you explain a bit more kind of what those technological advancements are? Do you think now looking 1015, 20 years into the future?

Sue Keay  06:10

Well, a lot of them are about automating a lot of the repetitive work that is currently being done in the economy and across a whole range of different jobs. So I'm sure I'm not the only person you know frustrated by the amount of email and calendar requests that you know I have to deal with on a day to day basis, that in many respects I feel get in the way of the work that I'm actually good at doing. And you know, ironically, a lot of these technologies are supposed to be helping save time. But since the advent of computers, you know, I haven't certainly noticed that there's been a huge increase in my leisure time. In fact, it's actually meant that I have to work harder to answer all of those emails and keep on top of the inquiry, since people are able to connect with me much more quickly and efficiently than they were in the past. So the opportunity with artificial intelligence is that, both at an individual level and at a company and governmental level, that they can be applied across a whole range of functions. And in fact, people are probably using a lot of these tools without even really being that aware of them. So anybody who runs a small business and has to operate their accounting software would probably be very grateful that they can now actually go through their bank statements and reconcile them with the accounting package, actually recommending and suggesting that this transaction looks like the one that needs to be reconciled with this part of the bank statement. I mean, that's a fantastic advance. That's just a very simple example. But you know, if you can start to scale a lot of those solutions, and in particular areas where we find it difficult to find people to do work, then that's obviously an advantage. And if we are able to implement the technology well, then one of the opportunities is potentially that we can provide services in areas that we struggle to do at the moment. An example of that would be delivering medical services to people in remote communities. So this is a real challenge. It's very difficult to get specialists to work out in those communities. And these technologies are starting to open up pathways that we can provide both prescription medication, you know, expert advice to people online in these communities that otherwise they might have to travel several hours to be able to obtain.

Neil Martin  08:30 

And I mentioned in my introduction physical and non-physical. It sounds a bit maybe confusing for people, but I guess that's AI versus robots. Would that be correct?

Sue Keay  08:42

Yes, it's a little bit broader than that, so you're right. Non-physical AI is anything that doesn't have physical embodiment, and so that could be anything that's on your computer. It could be an app. It's something that might be applied to cognitive work. Whereas physical AI might include robots, but it also might include sensors and other physical devices that are collecting information about the world around them and potentially making decisions based on that information. So, you know, an example of that might be some of the sensors that we have monitoring water flow in storm water drains, you know, to give some warning, if there's likely to be flooding. That all happens automatically, without people necessarily having to go and check flood gages.

Neil Martin  09:30 

We’ve had robots in factories for, you know, decades now. Do you see that just kind of increasing and increasing, or is it going to diversify and new technologies being created that aren't just kind of like churning out widgets, for example.

Sue Keay  09:47

Well, the traditional robots that we've had on factory floors for the last 70 years don't necessarily incorporate a lot of artificial intelligence. And so they're very good at repetitive tasks, but they've pretty much been pre-programmed to do what they have to do. They are not safe to work alongside humans - so if you go into a factory where they are deploying robots, you will usually see that they are caged off so that humans don't come into contact with them. And that technology has been, you know, very useful for decades now, but the new type of robots that are coming onto the market are much more flexible and able to work in unstructured environments that has not been possible for robotics in the past. So traditionally, you would only deploy robots in environments that you can control, either factories or warehouses, areas where perhaps the robot doesn't have to move from a single spot, but now the opportunity is that we can have mobile robots that are able to work side by side with humans, that don't have the same safety concerns, and that have much more utility, and because they're using artificial intelligence, much more capable of making decisions and changing their actions depending on the circumstances which your traditional industrial robot has been unable to do. I mean, unfortunately, and that's why we have to keep them caged, if you have a robot that's programmed to move something from A to B and a person gets in the way, the robot has no capacity to danger to stop. I mean, obviously we have a lot of safety protocols to try and make sure that happens, but they're not completely fail safe.

Neil Martin  11:21 

So I think what you were describing to my mind sounds like a very humanoid robot. And I seem to be seeing more and more on kind of social media companies displaying the acrobatics, for example, that a humanoid robot can do. It's kind of very unerring how human they're kind of now getting, is that correct? And I guess you talked a little bit about it there, the opportunities that that brings. It's also quite sci-fi-esque. Feels like something that I was watching on Star Trek, or something, you know, 20 years ago.

Sue Keay  11:56

Well, I think humanoid robots do make for great YouTube videos. I do think that people, when they get the opportunity to interact with humanoid robots, are likely to be very disappointed. The way that robots were evolving was more along the lines of mobile robots on wheels or on tracks that actually pretty stable and robust, whereas bipedal, two legged robots that have a humanoid form actually present a lot of fall risks, often very difficult to maintain their balance and to do useful tasks. There are a lot of problems that we haven't solved in terms of making those robots useful, for example, tactile sensing and gripping so that those robots can truly replicate what a human does without thinking every day, like opening a door, going up and down stairs, all of those sort of activities are still challenging for a robot, and something that you probably are less likely to see in the YouTube videos of them doing acrobats or boxing and things like that. If you go back only a decade ago, to develop a humanoid robot, was really economically out of reach for most people. So I would say in Australia, we would have been lucky to even have two humanoid robots in the country, and that's because they cost several million dollars. And of course, you didn't really want to get them to do many mundane tasks, because you didn't want them to fall over and need to be repaired. But you know, because of this huge investment that we're seeing in humanoid robots, we are now starting to see companies promising that they're going to be producing 10s of 1000s of these robots. We're starting to see more humanoid robots on Australian shores. They are getting to a price point where probably a lot of companies are going to experiment with these robots, but I think anyone who has worked in a robotics lab will probably counsel you to not have very high expectations.

Neil Martin  13:51 

I was going to ask, is it hype? And maybe two additional questions, when would you say that that is feasible, that that would happen? And second of all, could I get a humanoid robot in my house to do all the chores that I don't want to do, and then I can be more productive.

Sue Keay  14:05

Okay, so I don't think we will see a useful humanoid robot in the next decade. We will certainly see more humanoid robots, but I suspect that they will only be used for very specific tasks. But the big benefit of all of this investment into humanoid robots, and this is exactly what we've seen with self driving cars. So there was a huge wave of investment into self-driving cars, yet we still don't have a self driving car. But what it meant was that there were huge advances and also a huge reduction in the cost of many of the things that go into making a self driving car, so sensors, battery, technology, all of those things really shot off because of that huge investment. And what I hope is that we will see the same in humanoid robots. And in particular, I think what we will see are a lot of advances in what we call tactile sensors that could allow the robots to. Have hands that operate much more like a human hand would be able to operate, which is a particularly difficult problem in robotics. So you know, in many respects, it's going to lead to a lot of advances. Are you going to have a robot in your home? I certainly wouldn't have one in my home, and I would definitely not have one in a home that has more than one level. And that's because I struggle to see how we are ever going to make a humanoid robot functionally safe to operate in an environment with a family where you might have pets or young children, because when those things collapse, you know, they probably weigh a minimum of 20 kilos, and you have to be sensitive about what that means for anything that's going to be in their way if they fall down the stairs. How do you make sure they land safely? There are a lot of open questions around safety that are going to need to be addressed.

Neil Martin  15:53 

Maybe it sounds exactly like the self-driving cars. My understanding is that self driving cars are kind of fine if they're on a motorway and they're just going in a straight line, and there's nothing really in the way, but in the cities where you have people crossing the road randomly, or cars pulling out randomly, that's where they struggle. The environment makes it difficult for them to adapt and to and to operate properly.

Sue Keay  16:18

Well, exactly would you drive one out to Broken Hill. You know, there are so many obstacles that it would be hard to train those self-driving cars up to be prepared for any scenario. Now, obviously people are trying to do that. And in fact, you know, companies like Volvo, I believe, have, you know, actually developed a, you know, kangaroo detector for the Australian environment.

Neil Martin  16:41 

Country specific.

Sue Keay  16:42

Well, that's the sort of thing that you have to think of because, you know, I guess, as humans, one of the, you know, the advantages that we still have is that we're very adaptable and can change depending on the circumstances that are required - and that's still a big challenge for AI.

Neil Martin  16:58 

Do you think that it's valid that the public kind of do have some fears around AI. Maybe they might not completely understand it, fears about jobs, maybe?

Sue Keay  17:07

Yeah, I think Australia is interesting. And I don't think we have a great answer for why Australia tops those polls. It has been speculated that that is somehow related to Australians having a low level of literacy about artificial intelligence. But on the other hand, Australians in our personal lives are often very quick to take up new technologies, and that includes generative AI tools like ChatGPT. However, I do think that it's true that we haven't had a lot of national discussions about artificial intelligence and what it means. And in fact, I would argue that until the federal election earlier this year, you didn't ever really hear our government or our politicians talking about artificial intelligence. And so I think fundamentally, that holds us back from having a sort of mature, sophisticated view of the potential dangers, but also the potential opportunities of the technology. So I'm not quite sure that it's literacy. I think that we actually just need to be talking more about the subject and what it's you know, what it means for people. I think that it's also been a lot easier since the advent of ChatGPT for people to get a sense of what artificial intelligence is, because they can actually feel it and touch it themselves, right and play around with it. The disadvantage of that is that often people think that generative AI in the form of AI tools like ChatGPT, is the entirety of artificial intelligence, and it's definitely isn’t.

Neil Martin  18:42 

One of the things that I think maybe a top fear is jobs being made redundant by AI or robotics. I guess. Do you think that is a valid fear as well? Do you think jobs will inevitably be lost, and how do we deal with that?

Sue Keay  18:58

Yeah, I think there will be job displacement, which means that there will be jobs that will cease to exist, but there will be other jobs that will appear. And I think the uncertainty is that no one is very sure about what those new jobs might look like. And so I think it's completely understandable that people have concerns about that, and that is one of the reasons the Australian government has taken the position that they would really like to have unions involved in a lot of the discussions about how artificial intelligence is being deployed in the workforce. It also goes a lot to corporate responsibility to the workforce. I think one of the things that I would like to see that might go some way to allying people's fears is more serious effort being placed towards looking at workforce transformation and what that actually means, what the workforce of the future is likely to look like? What are the skills that people are going to need to have? How do we make sure that people can get those skills? We need to be having far more of those conversations, and I think that both government and industry need to be investing much more time in mapping out what that future might look like, to give people that sense of confidence that they're not necessarily going to be displaced. Unfortunately, unless we can raise our productivity, some of these industries will continue to move offshore, and so you have to have a balance. It really isn't an option for us to just bury our heads in the sand and hope that this technological change won't impact on us. We have to figure out how we can harness the benefits and make sure that we still have a lot of these industries onshore.

Neil Martin  20:37 

I guess the ultimate kind of I don't know whether it's dystopia, but if people keep losing their job, if people did lose their jobs, and that process carried on, you would start to wonder, how are people going to afford to buy the things that these robots and AI are creating? So it becomes maybe a social science question or an economics question, but you're kind of at the start of that process, I guess.

Sue Keay  21:03

I think the biggest challenge is how we can make sure that we are extracting value here in Australia, because artificial intelligence, probably possibly more than any other transformation in the past, has the very high potential for being an extractive industry. And the people who are leading the development of these tools and making all of the profits from them are not Australian companies. So obviously, you know, we are seeing a situation where companies are always worried about shareholder value, looking at ways to get around the tax system, going to lower tax rate countries where they feel that they are not getting a good deal from a country, and at some point that has to stop. And you know, I'm not, you know economist or, you know, a policy person, but I do wonder whether corporations law is not fit for purpose anymore.

Neil Martin  21:58 

I can take you back to something you mentioned there before, I think talking about sovereign capability, especially where AI is concerned. Can you explain a bit more about why you think it is so important that Australia has sovereign capability in that space?

Sue Keay  22:13

I think anyone who has used ChatGPT and had their spelling corrected into US English, or, you know, receives answers that just don't really resonate for you as an Australian, would appreciate that it would be fantastic if we had some of these tools that actually were more reflective of our culture and values. And I think there are some really good examples of where it's essential that we be developing our own artificial intelligence. I mean, just look at the Australian population. We are unique compared to the rest of the world. If we are developing artificial intelligence for health, for example, basing that on a population from the middle of the US is likely to have very little in common with a population that you would get from Australia. We have a lot of valuable data sets in areas like health and in some of our key industries. So for defense, mining, energy, agriculture: these are all areas where it's really crucial that we be developing our own Australian AI. And I guess, you know, at the end of the day, national security. There are a lot of things, you know, for example, our banking and financial services, where it's important that we keep all of our data on shore, and a lot of the competitive advantage that those industries have need to be restricted to Australian entities, not so that other countries can figure out how they can kind of game the system and extract value from Australia without really giving us anything in return.

Neil Martin  23:47 

Yeah, and I think we've spoken to other AI experts previously on other episodes, and I think there was a conversation at times about who's controlling the information on the back end? Who's kind of writing those algorithms that you can't control as the user? That kind of would be important, potentially as well, to have that being based from Australia.

Sue Keay  24:11

Yes, definitely. And I hope that it's something that our government and our national security community are taking very seriously, but I also think that it's something that we as individuals need to be more cognizant of because we've been conditioned to think that it is acceptable to give up a lot of our personal data for convenience sake. And I think we have to stop unquestionably accepting that this kind of one way flow of information, which is very extractive, is actually acceptable, and if we have more control of our data, then that lends itself for us to be able to build our own AI specific models that are beneficial to us and not necessarily to anyone else. So it's really critical, but I think I would encourage everybody to think about all of the data that they're currently giving away for free, and how much value these tech companies are extracting out of that, and at some point we have to say: ‘No, you can't have my data for free.’

Neil Martin  25:07 

In practice, how do you think that that would work? Would it be driven by the government in terms of policy? 

Sue Keay  25:13

Well I’m sure the government wouldn't like hearing me say this, but I do think there's a place for policy and regulation. And I'd have to say I think our E-Safety Commissioner is doing an absolutely fantastic job at looking at trying to protect the Australian public from the harms of much of this technology, and we've seen that with the social media ban. I'm really hoping they're going to act very quickly to prevent young adults and children from being able to access things like synthetic relationships and AI companions, which have a high potential for psychological harm. I mean, arguably, we shouldn't be exposing any member of the Australian public to some of these untested AI tools. So, on the one hand, I think you know, we are doing as much as we can in some of that space, but you know, there are other areas. And going back to the physical AI, you might have heard of a robot called Neo, which is a household robot that can learn how to do tasks. And at the moment, it's hideously slow. It might take six hours for it to stack your dishwasher, but they are selling this robot for, I think, about $20,000 US dollars into households, and it can literally do nothing. What the robot does is it's teleoperated, so there is a human operator 1000s of kilometers potentially away, who is looking into your home and figuring out how to do tasks, and that's because these companies need that information to train up their robots so that then they can actually have a commercial product. But I can't understand why anyone would pay for this. It is, it is a terrible idea. If the companies want to get that sort of data, they should be paying you a couple $100,000 for, you know, one, invading your privacy, but two, having you train their systems up. But I think that we have become quite unquestioning about this idea that we give up our data and privacy and we get some, maybe some cool, convenient stuff in return.

Neil Martin  27:17 

So if I look to the future, and I project, say, 20 or 30 years in the future with the AI and the robotics in terms of increased productivity, if Australia gets all of the things right that you've mentioned, what do you think that ends up looking like?

Sue Keay  27:35

I would hope that what that looks like is that, you know, we're producing a lot of our food untouched by human hands. So it's not that we don't have human farmers, but that we take a lot of the load away from them by giving them the tools to be able to grow our crops, to herd livestock without having to be there every day and, you know, go through a lot of the hardships associated with that remote, sometimes very lonely work in mining. I think we will start to see far more robots out on mine sites. At the moment, you tend to just see those big yellow robot mining trucks on iron ore sites, but I think there's high potential for there to be a lot more of that, and we know that there are a lot of particularly mental health issues with fly in, fly out work as well, as you know, some challenges in terms of what that means for remote communities. So, I think that those are some things that we can do. I think, as I previewed earlier, I think there are a lot of opportunities for us to be able to deliver services in remote areas, which we have not been able to do before, and help potentially reduce the gap between, you know, the healthcare services available for people in the city versus people in the country. And you know, I think it's critical for our defence. So, you know, there are a lot of things where we just need to be good at this, because it's a global competition, and we can't be powerless or give up our agency to other countries unless we're prepared to potentially lose our sovereignty.

Neil Martin  29:13 

And I guess, on the flip side, and maybe without being too negative, if we didn't take up all of these opportunities, what would you be fearing would be the consequences? What's the flip side of that kind of positive outcome?

Sue Keay  29:27

Well, the flip side is that we become completely dependent on, you know, at the moment, what we would be dependent on is the US tech companies for our future. And they don't necessarily have to provide us with these services, and I'm not sure that you could necessarily call them our friends. You know, they have commercial interests that they have to meet. They have shareholder obligations. What are their obligations to the Australian public?

Neil Martin  29:55 

And you might have mentioned this before, but who do you think needs. Be driving this forward, and maybe, how should it be driven forward?

Sue Keay  30:05

Well, I think we're lucky to live in a country where, despite the fact that we put up a lot of obstacles in their path, on a constant basis, I am seeing people who are starting up AI and robotics companies. So we're clearly a nation full of optimists, because it's definitely not easy to do that on our shores. And I guess, by benefit of my role with UNSW, I'm really fortunate, and I'm sure you feel this same way, to see the amazing talent that we have. And you know, the, I guess, the dreamers, the visionaries who are creating, you know, fantastic new technologies. We just need to find a way to harness that. And the responsibility for that doesn't just lie on the government. I mean, I think there's a number of things that we need to look at in terms of, you know, being able to provide more capital to see, you know, these opportunities come through to fruition. But I also think there is a role for government in really taking a leadership position and having some ambition in this space. So, I often compare us to a country like South Korea. So our population, I think, is 28 million. South Korea's population is 51 million. So they are a larger country, but their GDP is the same as Australia, or just a bit smaller, and they are investing really heavily in artificial intelligence, and that is because back in 2017 they put out a national AI strategy with the ambition to be the number three country in the world in AI and by many metrics, you'd have to say, over the last eight years, they've succeeded in many of their ambitions, and that's because if you don't know where you're going, then any road will take you there. So, what I would love to see is that we actually have a clear idea of where Australia wants to be, where artificial intelligence is concerned, and I think we should be ambitious. We've got the talent, we've got a lot of the technologies. We just have to figure out ways that we can ensure that the benefits of all of that accrue here in Australia, and that the value is not being extracted and taken offshore.

Neil Martin  32:07 

And one of the things finished, maybe I should have mentioned this earlier, or asked it earlier, but I see some things online more and more about kind of AI eating itself. Is that a hype? You talk about there was some hype? 

Sue Keay  32:21

Not at all.

Neil Martin  32:22 

What would you say to people who I read those articles online, I think they talk about cannibalism, kind of taking its own output, and you talk about hallucinations, where it comes up with things out of thin air. There are some negative comments, I think, about AI maybe being a bubble that's going to burst. What would you say, I guess, in response to that?

Sue Keay  32:44

That is definitely true of these large frontier models that underpin the generative AI tools like ChatGPT. 

Neil Martin  32:52

And what do you mean by frontier model?

Sue Keay  32:54

Frontier models are these large-scale models that have essentially been trained on everything on the internet, and that require enormous amounts of data and computational power to be able to produce. So, you know, the main frontier models are very much only from US or China. And so that powers things like ChatGPT, and in the case of China, DeepSeek. But the problem with that is that they've run out of data to steal from everyone, right? And so now, because there's so much AI slob being generated and then being put in the web, if you're training things now off that, without thinking carefully about how you're curating the data that goes into those models, then it is going to be curating all of that AI slob. We're all being directed towards there being one answer to every question, and that's just not healthy. And that's exactly what they're talking about when they're talking about the cannibalisation, because the AI models, if they're trained up on AI material, they're just going to lose any relevance at all, really. And I think that that's another reason why we should be developing our own AI because people think that these frontier models are almost the only artificial intelligence around, and that's just not true. We have so many opportunities to be developing domain specific models for things like mining, agriculture, health here in Australia that are on curated data sets, that are respectful of copyright holders ownership, and that, you know, can have far more benefits than these huge frontier models with their hallucinations and where you can't even interrogate to find the original source or really understand - they're just black boxes. We can do so much better than that. And you know, I'm really looking forward to when that happens. 

Neil Martin  34:43

Me too, and this whole episode, I think, has made me look forward to what the future holds. Dr Sue Keay, once again, many thanks for joining me and sharing your thoughts and opinions.

Sue Keay  34:53

Thanks, Neil. 

Neil Martin  34:54

I guess we'll just have to wait and see whether I get replaced by an AI powered humanoid robot for the next episode. Unfortunately, that's all we've got time for right now. Thank you for listening. I've been Neil Martin, and I hope you'll join me again soon for the next edition of our Engineering the Future series.

Outro  35:15

You've been listening to UNSW Engineering the Future podcast, highlighting the people and ideas, creating real world impact and enabling progress for all. Don't forget to subscribe to stay updated on upcoming episodes. Check out our show notes for details on in person, events, panel discussions and more fascinating insights into the future of engineering.