AI at work: Professor Toby Walsh on the skills AI won’t replace

About the episode

“The real threat isn’t a machine replacing you... it’s a competitor using AI to outpace you.”

If you’ve been weighing up how to bring AI into your workplace but are hesitant on how to do it, you aren’t alone.

Australia’s leading AI scientist, Scientia Professor Toby Walsh, Chief Scientist at the UNSW AI Institute says that, by handing AI your dull, repetitive tasks, you can focus on high-level decisions that actually drive growth.

Learn how to invest in an AI-literate workforce and why your human skills are about to become your most valuable asset.

Interested in AI disrupting other industries? Listen to our episode with UNSW alum Jackie Rabec, co-creator of ThinkMD ai, an app revolutionising healthcare.

Want to know more? 

Want to follow the latest research and news from UNSW Business School and AGSM @ UNSW Business School? Subscribe to BusinessThink and follow UNSW Business School and AGSM on LinkedIn. Or listen to more episodes of The Business Of and subscribe to our UNSW Business School and AGSM channels on YouTube.

Transcript

Scientia Professor Toby Walsh: AI is a wonderful technology, but you're the best people to solve your business problems. I always say to people, there's no AI problem, there's a business problem, and AI might be a tool that helps you solve that AI problem. And who understands what the business problems are? Well, it's your employees who understand exactly where the pain and pressure are and where the opportunity for innovation is in your business. So you need to empower those people, give them the AI tools so that they can go off and make your business a better business.

Dr Juliet Bourke: Have you thought about how AI is reshaping your job? AI is now a billion-dollar-a-day industry, and it's growing faster than we can adapt. The impact is already visible across the global tech sector. Companies that built the digital economy are now rebuilding around AI, cutting jobs and shrinking teams in Australia; some of the biggest players in software are already making moves wisely. Tech Atlassian, Block Inc and Salesforce are all restructuring with AI at the core. Is this just an industry reshuffle, or is it signalling the biggest shift in work as we know it? And as a business leader, how can you adapt to make yourself future-proof? 

Hi, I'm Dr Juliet Burke, adjunct professor at the School of Management and Governance, and this is The Business Of a podcast by UNSW Business School Scientia Professor Toby Walsh says that up to a third of Australian workers could be disrupted by 2030. He's one of the country's leading AI experts and chief scientist at UNSW AI Institute. Toby, over the last year, the tech industry has seen mass layoffs, hiring freezes and a shift to smaller AI-driven teams. Is big tech the canary in the coal mine here? Or is this already the beginning of a wider workforce shift?

Scientia Professor Toby Walsh: I think it's going to be a significant challenge to large legacy businesses, because they're not going to be as agile, and they won't be able to move as fast. I mean, that's the thing about AI at every level: you've got to be agile, and small companies that can use these agentic tools don't need to hire a marketing person. You can use a marketing agent to do the marketing for you. You don't need an HR person. You could have an HR agent who does the onboarding of new people. So I think it is going to challenge those larger businesses, but on the other hand, you have the data, and you have the IT expertise. That is going to mean that, if you put your mind to it, I think artificial intelligence is going to be as transformative for your business it is for these new startups.

Dr Juliet Bourke: Okay? And I'm interested in those larger businesses; they seem to be going through this phase of layoffs and cost-cutting, and I'm wondering if they're doing that because the return on investment for AI just hasn't been there, and they're actually rerouting some money in a different direction.

Scientia Professor Toby Walsh: There's some of that. Certainly, there are spectacular, large investments being made, and we're not seeing the returns yet. So they have to look at where to get that money from and where to make savings to be able to afford to make those very large CapEx investments. Secondly, I think you've got to be aware that there's a bit of AI washing where a company that's been perhaps not very well managed, and rather than say that we've been poorly managed, which doesn't do much to your share price. They've been saying, Well, we're investing in AI, and then your share price goes up. So the CEO, he or she looks good if you say it's AI, you're making 10% layoffs of your staff, as opposed to admitting, well, actually, it was my fault, and we've been very well managed.

Dr Juliet Bourke: All right. Well, let's just have a little look at this sort of move from impact to action. And I'd love to understand a bit more what it looks like, actually, in a business. I understand that AI can already perform tasks at roughly 10 to 15% of labour market input in some sectors. Is that about right? That number?

Scientia Professor Toby Walsh: I mean, economists always come up with a number that's around about 15% for the potential savings. It's not uniform. And certainly, you know, it's not surprising that the language models we have today are conversational tours. They have a good command of language. They are so they're very good at conversational things. So that's things like customer services and marketing, those sorts of things. Of course. You know, the next wave has been these agentic tours, which are, instead of having chatbots that sit there and wait to answer your questions and do things according to your bidding. They can break tasks down into a bunch of goals and solve those things. For example, you get a new employee. There are a bunch of things you have to do. You have to register them with finance so they can get their salary paid. You've got to get them a security card. You've got to do the various health and safety onboarding, and so an agent can go through all those steps and seamlessly, hopefully onboard a new person, a member of staff. That's good news for the business. Perhaps not such good news if you're someone who works in HR, because a big chunk of what you do is now being done by one of these AI agents.

Dr Juliet Bourke: Is it more at the beginning of a career that the jobs are changing, or is it certain industries?

Scientia Professor Toby Walsh: That's a great question. Yes, it's certainly the case today, the sorts of capabilities that the AI models have, the sorts of things that people do when they start in their career, graduate-level or entry-level jobs, those are the sorts of things that AI seems to be able to do very well. Today, there's data showing that, you know, there's been a decline in graduate level recruitment charts in the FT the other day, showing in the UK and the US, and I suspect, very similar numbers in Australia, 30% decline in graduate level job adverts in the last three years. That's a significant change, and you're certainly, certainly, lots of anecdotal evidence. I'm certainly hearing from students at the university that they're finding it much harder to get their foot in the door. The problem with that is that maybe it makes sense for that business not to hire those people, because what they can do is now being done by AI. But then you have to think a longer term. What's going to happen in the long term for those businesses? Businesses are most business organisational structures are pyramids. A lot of people go into the bottom, and fewer and fewer people rise up through the ranks and become managers and so on. So where are those middle managers and higher managers going to come? Where are they going to learn about the business when you become a, you know, first-year lawyer? You spend a lot of time studying the nuts and bolts of contracts, so you become an expert on contract law. Maybe we can get AI now to write most of those contracts. They're quite formulaic, and AI is pretty good at doing those sorts of formulaic things. But how do you ever learn about contract law if you haven't done the 1000, 10,000 hours, whatever it is, that's required to become an expert? So I think businesses are going to have to wake up to the idea that, yes, in the short term, maybe a lot of this stuff now is being done by generative. AI for you. But where is that going to leave your business? The most valuable thing that a business has is people, at the end of the day, it should be investing in those people, and maybe you're going to have to spend money on people doing things that aren't strictly necessary, because otherwise those people aren't going to have the skills you need in the future.

Dr Juliet Bourke: And talking about those skills, I mean, of course, the technical skills are important at the beginning, and that's what you just talked about, learning the craft, but I wonder if people are also learning at the same time, skills of judgement, social interaction, communication, empathy, negotiation, and where's that going to come from if they don't have that first five years in the workplace?

Scientia Professor Toby Walsh: Yeah, that's the human advantage. Those are the things that we should be investing in. Those are the things that distinguish us from the machines. Those are things I think that, however capable, however smart the machines become, humans are always going to have the edge. And so those are the things that certainly, I highly advise people, if you want to be investing in yourself or your employees, those are the sorts of skills that are going to be ever more valuable, ever more important, because those are things that artificial intelligence is not going to bring to the table.

Dr Juliet Bourke: And so if someone is then five to 10 years in their career, is it about learning the technical skills of AI and really doubling down on human skills?

Scientia Professor Toby Walsh: There's a challenge for people in the businesses, which is that they have to be more digitally native than perhaps they are. Their challenge is not that they don't have the grounding in the business, the understanding and those, and those, those social and emotional intelligence skills that you mentioned. Their challenge is that the technology is moving very fast and much faster than they're perhaps comfortable with. There's quite a bit of evidence, whether it be for doing law matters or computer programming, that the tools, the AI tools that we have today, amplify the differences between people that that people who aren't very good, if you're not a very good programmer, the AI tools, AI programming tools that we have, like coding tools, actually expose your weaknesses because you spend more time debugging the code, because the code is not perfect. AI tools are pretty good, but they're not perfect. You spend more time finding where it's put a few bugs in than you would if you programmed it yourself without any tools. Whereas, if you're a really competent programmer, then the AI tools really double and triple the speed at which you can write code.

Dr Juliet Bourke: When you read something that AI has generated, and you actually know what you're talking about. I heard this term work slop. You know that it has produced work slop, but someone who's not...

Scientia Professor Toby Walsh: Please, anyone listening, don't write your LinkedIn posts with AI. I can't stand reading AI because it has a particular voice. You can recognise the voice. It has some style tics that you really easily recognise. It's not just the em dash, it's the tricolon. It always uses threes. That's a, you know, old rhetorical trick. It's always using contrastive phrasing. It's not this, it's that. It becomes very cliched. And once you know, I've spent my time, unfortunately, reading too much AI-generated content. And it just so annoys me if you can't be bothered to tell me what you think I'm not interested in, what AI thinks I'm interested in, what you think I'm interested in, the human perspective, new novel perspective. Because this is how the AI tools work, they're giving you a very competent, average, middle-of-the-road answer. They've been trained to be that way. If I want to read an interesting point on LinkedIn and an interesting post, I want someone who's going to take an interesting angle, not something that's going to be very average.

Dr Juliet Bourke: Do you think we've gotten smarter about, as you said, picking out that that's been written by AI, I'm actually not interested.

Scientia Professor Toby Walsh: No. I mean, I actually unfollow those people? 

Dr Juliet Bourke: Right,

Scientia Professor Toby Walsh: Yeah. No, we have got smarter, although I have to tell you, we've got more stupid.

Dr Juliet Bourke: And what does that mean?

Scientia Professor Toby Walsh: Well, because there's this well-known effect, the Flynn effect, which is, over the last century, people on average have got smarter. There's one or two percentage points in IQ every generation, which is fantastic. There's better education, better diet, various things that have contributed to the fact that slowly but surely, humanity as a whole has been getting smarter. Well documented in many countries. But unfortunately, in the last decade, there's been a significant downturn. We are actually now getting more stupid. It's very hard to measure this. It's very hard to actually sample intelligence very well. It's done best in Scandinavian and other countries that still have conscription, because what happens is every 18-year-old is conscripted into the army, and then they do a uniform intelligence test to work out who's going to be an officer and who's going to be a private. And so you get a very good sample, an unbiased sample of the population and their IQ. And in those Scandinavian countries where that's happening, or a place where conscription still measures IQ, we've seen a one or two percentage point decline in intelligence as measured by IQ tests. We're also seeing in other measures, in terms of attention span, famously going down. Kindle measures this way: how long people read for and how long before they skip out of reading a book. So by various measures, we seem to be getting more stupid. I always joke that the machines will become smarter than us because, not because the machines get any smarter because we got more stupid. But there's also evidence that unless you put the effort in, you need the grit, you don't learn. Yeah, the thinking your brain is a muscle, yeah? And if you haven't put the effort in, if you outsource too much to the tools, then sure, you get those things done, but you never actually stretch yourself. You never actually learn.

Dr Juliet Bourke: Just thinking about then, you know, companies themselves, perhaps government from a policy perspective, what is it that we should be doing as business leaders to mitigate against these unknown risks?

Scientia Professor Toby Walsh: Well, the good thing here is, I think companies are starting to wake up to the idea that there's actually going to be a competitive advantage to be responsible in the way that we use artificial intelligence. Being responsible is actually going to give you a business advantage. Your customers are going to recognise that you're doing the right thing. I mean, I perhaps name a few businesses here, actually, you know, why is it that there's a dislike in many courses for companies like Facebook Meta, but there isn't a similar dislike for someone like Amazon. I mean, actually, when you look at what the businesses is, both are data Hawks. Both businesses are built upon taking our data and selling our stuff based on that data. In the case of Facebook, I think it's the case that people realise, well, what am I getting in return? You know, I'm getting a lot of negatives. I'm not getting so many positive. Case of Amazon. We're getting, you know, next-day delivery, or even same-day delivery. Today, we're getting probably the best prices we could find, even if we shopped around. So although they're using, collecting a lot of data and using that data to sell us stuff and so on, I think most people feel, well, I'm getting a reasonable exchange in terms of Amazon's been reasonably responsible. Why is it we like Apple? Because Apple fights really hard to protect our privacy, they will actually resist federal lawsuits where they've come across a device connected to someone with a terrorist activity. It's locked. It's an Apple device. Apple has refused to unlock the device.

Dr Juliet Bourke: I just want to pivot slightly and just think about AI on a sort of day-to-day basis, and I'm wondering what that's doing for the role of employers and employees, just at a very hands-on level.

Scientia Professor Toby Walsh: There's a lot of shadow AI in many businesses, which is that many businesses have been, perhaps, slower to adopt. AI, their employees haven't been slow to adopt AI. I mean, that's the fantastic thing about the technology. It's very accessible, and lots of people have just started playing with it and have started using it to help them prepare draft reports or write emails and briefs. It's a fantastic tool for brainstorming ideas as well. So lots of people have already integrated it into their life, even though maybe it's not actually part of the, you know, the corporate IT structure, or even even possibly allowed now that is, carries a significant number of risks, one of which is that people may be typing into these tools highly confidential, sensitive data, which they're sharing now with Microsoft or open AI, or whoever it is, you have no idea where that data is going to end up. It's going to probably be used to train the next iteration of the tools.

Dr Juliet Bourke: Don't think people realise that when they're using chat, GPT, the free version, that that data is going to go somewhere?

Scientia Professor Toby Walsh: Well, maybe they do, but they think it probably is not. No one's going to notice. And it sets the tool so useful, I couldn't live without it now. So this is why you need a corporate policy, and you need governance.

Scientia Professor Toby Walsh: You need governance, and you need to have a contract with a company so you can say what's happening to the data. You can actually say it's been sandboxed, it's put in a safe place, so that you can turn to your customers and say to all your clients and say to them, yes, your data hasn't left the building. We've, you know, we've actually been careful guardians of all that. So yes, that's something you have to be very mindful of. But equally, it's not just a case of stopping that, you know, issuing a proclamation that your business needs to stop, that people need to stop doing that. It's to wake up to, why are people doing that? There's obviously a demand. You need to satisfy the demand because people are finding the tools useful. So you've got to find a way that people can use the tools and not expose you to those issues of privacy and so on.

Dr Juliet Bourke: Could you just talk me through Toby Agentic AI? What does it mean?

Scientia Professor Toby Walsh: Agentic AI is the latest wave of artificial intelligence. We had generative AI language models like ChatGPT, which could answer questions, summarise and synthesise information. Now they're moving into something much more useful in many respects, which is to rather sit there and wait to answer your questions. They have some agency, some ability to go off and do things. They can take a problem and break it down into subtasks, and then break their subtasks into further subtasks and do those things. So with agentic AI, people can set really long-term goals, things that go over time, or things that go forever, which is like, you know, here's an agent that's in charge of my diary. The agent is going to be given various rules as to, you know, setting up meetings with various people who to give priority to which things you're allowed to move around, which things not to allow to move around. So, yes, it can give, be given much more executive authority, and it can work on much greater timescales. You know, a good way to imagine the future. This is Hal Varian, the chief economist at Google, which is what rich people have today. So rich people have personal assistants. They go off and manage things for them. They book hairdressers and restaurants for us, and manage our diary and do all that day-to-day stuff. Well, we're going to have an agentic assistant that does exactly that, that learns all our preferences, learns how to manage our diary, learns, you know, all the things, remembers to order flowers for our partner and on our anniversary and so on. And similarly, like rich people, have personal bankers who manage their property portfolio and their investments and so on, and monitor them and move things money around when needed, when markets seem to be moving well, well, suddenly we haven't. We have a, you know, agentic personal banker that manages our much more modest finances and does those sorts of things for us.

Dr Juliet Bourke: What's your take on AI cloning for people at the bottom of an organisation? But also CEOs, I've heard of Trinny Woodall. She's the CEO of Trinny London, and she's creating a virtual clone. What do you think?

Scientia Professor Toby Walsh: The whole risk that we have with these agentic tools, which is that increasingly they're not just answering our questions, but they're actually being given the responsibility to do stuff, to turn up to meetings and make executive decisions, to spend money, to contact customers, to make replies to sensitive customer communications that come with much greater risk. I mean, if you thought that AI introduced risk, you wait till you think about the risk that agentic AI introduces, where things that humans used to have oversight of and responsibility for, that's the most important thing, which is that someone's going to have to take the buck for when things go wrong and when those decisions are bad, decisions, poor decisions, decisions that you get sued for. And that I can promise you, the people who are going to be sued, the people are going to have to take the fall, are the humans, not the machines.

Dr Juliet Bourke: And so what will be the consequence for us as humans when we have these agents doing? These tasks that are, you know, long, complex, boring tasks. Well, what do we do as humans?

Scientia Professor Toby Walsh: Well, in a good world, we sit around and make art and read books.

Dr Juliet Bourke: Do we actually work less?

Scientia Professor Toby Walsh: That's one possible future. It is entirely possible that we work less, that we actually say, yes, okay, we don't have to do all those things that we used to do, because AI is doing a big chunk of them, and we can actually work less. And that will also solve some of the other challenges that people worry about. We talked at the start about how many entry-level jobs were disappearing. Well, maybe there are fewer jobs if we work less, so we can spread the fewer jobs around more people. So that will be one more solution to it, and people forget the weekend was an invention of the Industrial Revolution. We didn't have the weekend before. The benefits of industrialisation that we got with the industrial revolution, we humans came up with the idea as a gift from the technological improvements, the efficiencies that we got from automating our work. Well now, as we automate some of our cognitive work, maybe we'll get similar savings, or we could say, well, actually, we'll have another day off. And I mean, there's interesting, there's lots of studies now starting to happen on a four-day work week, and all of the studies uncover two surprising results. First of all, people are just as productive in four days of work as they were in five. So you can pay people just as much because they're doing the producing as much, right? And secondly, again, a surprising discovery, people are happier. Who would you imagine? You do less work. You spend more time with your family. You spend more time on your hobbies. No one gets to the end of their life on their final moments and says, "You know what, darling, I should spend more time in the office."

Dr Juliet Bourke: All right. Well, you've convinced me. I'm very happy. I'm just wondering, then, what is my next step in the direction of making that a reality in terms of AI and Generative AI, Agentic AI? What do I need to do as a business leader to get the best out of these things?

Scientia Professor Toby Walsh: Well, you need to lead. You need to show leadership. You need to make these tools something that is integral to your business. I mean, at the end of the day, the people who work for you are the most valuable thing you've got in your business, their knowledge, their expertise. That's the thing that you should be investing in. You should be investing in them becoming AI-capable, AI-literate people. And the great thing is, it's actually quite easy to find lots of AI tools that are many of them free or very cheap. There are lots of training materials available these days to actually get up to speed. It's, of course, a very fast-moving field, so it's a constant thing to keep up to speed these days. But there are fantastic resources and opportunities to say, Okay, let's try this out. I mean, obviously, don't put it in something that's mission-critical. Not all AI projects succeed. Many of them fail. You've got to be very agile, but start investing in your people. Start encouraging them, providing incentives for them to pick up, and you start using the tools, because AI is a wonderful technology, but you're the best people to solve your business problems. I always say to people, there's no AI problem, there's a business problem, and AI might be a capability, a tool, that helps you solve that AI problem. And who understands what the business problems are? Well, it's your employees, the people who are at the cutting face of your business, who understand exactly where the pain and pressure is and where the opportunity for innovation is in your business. So you need to empower those people, give them the AI tools so they can go off and make your business a better business.

Dr Juliet Bourke: And is that the five to 10 year future we are just working in, better businesses, more productive businesses? We've got our three-day weekend.

Scientia Professor Toby Walsh: Well, yeah, hopefully it's that we're doing less of the four DS, the dull, the difficult, the dirty and the dangerous. Of course, the first place to look for AI to help you is in doing all those dull, repetitive things, because AI is going to be very good at doing dull, repetitive things. That's what it's very good at, and freeing the humans up. I always say to people, you know, you can see this as an opportunity. You can see a 10-15% efficiency improvement possibility here. If you start using AI to help make better decisions as a business leader, you've got a choice now. You could say, Okay, we're going to cut our workforce by 10%, which is an entirely reasonable, rational economic return from that. Or you can say, Well, okay, we've got 10% of these people's time freed up. We can now put those people to improving our product, improving our service, talking to our customers, and understanding better. I think if you want to be a business in the long term, probably the second is the better strategy, the one that you say, okay, we can lift our game because your competitors will. Some of your competitors may be lifting your game, and people are going to eventually notice. So it's an opportunity, yes, to reduce your workforce by 10%, or it's an opportunity to say we've got 10% more time to improve our business.

Dr Juliet Bourke: That's the end here. Professor Toby Walsh, Chief Scientist at the UNSW AI Institute. If you enjoyed this episode, we'd appreciate it if you left us a review. It helps more people find the show.

Dr Juliet Bourke: And if you're interested in how AI is reshaping industries in different ways, listen to our episode with Jackie Rabec. She's the co-founder of medtech company, think MD, AI. They're using AI to revolutionise healthcare.

Dr Jackie Rabec: The reality is that there will be facets of healthcare delivery that will be automated. In some parts of the healthcare system, it'll replace doctors, and in others it won't. And so that's something that I think we'll kind of have to navigate as a fraternity.

Dr Juliet Bourke: You'll find the link in the episode description. The Business Of is brought to you by the UNSW Business School, produced with Deadset studios.

To stay up-to-date with our latest podcasts as well as the latest insights and thought leadership from the Business School, subscribe to BusinessThink.