Episode 399: The Real AI Risks with Craig Unsworth and Dominic Bowen

In Episode 399 of The International Risk Podcast, Dominic and Craig examine the real economics of artificial intelligence: how CFOs should measure AI returns, when automation becomes job displacement, why agentic workflows are changing competitive advantage, and how AI is becoming a geopolitical and governance challenge.

They discuss:
– How CFOs can distinguish durable AI value from hype
– Why some companies are cutting headcount while others avoid new hiring
– The shift from AI experimentation to operational agentic workflows
– Competition and regulation across the United States, Europe and China
– How professionals can future-proof their careers
– Mass employment change, taxation and pressure on the middle class
– Who captures the profits when machines replace human work
– Shrinking business moats, talent risk and AI dependency
– Defence, data centres, energy, water and international risk

What concerns you most: job displacement, concentrated profits, weak governance or geopolitical competition? Share your view in the comments.

The International Risk Podcast brings you conversations with global experts, frontline practitioners, and senior decision-makers who are shaping how we understand and respond to international risk. From geopolitical instability and organised crime to cybersecurity threats and hybrid warfare, each episode explores the forces transforming our world and what smart leaders must do to navigate them. Whether you’re a board member, policymaker, or risk professional, The International Risk Podcast delivers actionable insights, sharp analysis, and real-world stories that matter.

Craig Unsworth is a Portfolio Chief Product Officer and senior adviser who works with private-equity firms and their portfolio companies. His work focuses on product strategy, transformation, growth and agentic AI.

Dominic Bowen is the host of The International Risk Podcast and Europe’s leading expert on international risk and crisis management. As Head of Strategic Advisory and Partner at one of Europe’s leading risk management consulting firms, Dominic advises CEOs, boards, and senior executives across the continent on how to prepare for uncertainty and act with intent. He has spent decades working in war zones, advising multinational companies, and supporting Europe’s business leaders. Dominic is the go-to business advisor for leaders navigating risk, crisis, and strategy; trusted for his clarity, calmness under pressure, and ability to turn volatility into competitive advantage. Dominic equips today’s business leaders with the insight and confidence to lead through disruption and deliver sustained strategic advantage.
Subscribe for expert conversations on geopolitics, security, economics, conflict, resilience and global affairs.

Podcast Chapters

00:00 AI value, jobs and risk
00:22 How CFOs should measure AI returns
01:50 AI productivity, layoffs and fewer new hires
04:14 From experiments to agentic workflows
06:14 AI geopolitics: America, Europe and China
08:39 How to future-proof your career
11:07 The AI tools and workflows leaders overlook
14:57 The cost of waiting to adopt AI
18:47 Who captures the gains from automation?
23:14 Who should govern AI?
23:34 Shrinking business moats and hidden risks
26:27 Defence, resources and data-centre risk
29:08 The leadership skills businesses need now
29:46 Closing thoughts

#ArtificialIntelligence #FutureOfWork #Geopolitics #AI

Episode transcript

Episode 399: AI will change every job: who wins, who loses?

Host: Dominic Bowen
Guest: Craig Unsworth

Dominic Bowen speaks with portfolio chief product officer and senior adviser Craig Unsworth about where artificial intelligence is already creating measurable value, how AI will reshape employment and taxation, and why computing power, data and infrastructure are becoming geopolitical assets.

Introduction

Dominic Bowen: And in this episode, I’m speaking with Craig Unsworth about where AI is delivering real value already and where that hype is outrunning reality and what leaders need to do now to stay competitive. Craig, welcome to the International Risk Podcast.

Craig Unsworth: Thank you for having me. It’s great to be here.

Dominic Bowen: So where should CFOs be looking for that first proof that AI is moving from this excitement that we’re all in to something that’s durable economically? Is it revenue? Is it margin expansion? Is it labour productivity? Is it customer retention? What are those first proofs that the CFOs should be looking for today then?

Craig Unsworth: I think it’s all of the above, which is not a cop-out of an answer. I think it really has to be all of the above. You have lots of options of how to deploy capital. You have lots of options of how to report and strengthen margin. It all starts with adding to revenue—the top line—and it continues through to making your operation more efficient. I think all of these things lead to key metrics, which every CFO will be watching obsessively just now. I think one of the most interesting things is the direct impact on people. And one of the metrics that I’m using now is FTE-E, so the full-time equivalent equivalent. So literally taking a list of people, human people, and saying this is what this number of people used to do. And now this is what people plus agents, people plus tooling, et cetera, can do instead. The delta being the FTE, and that, for as long as that number is larger than the additional spend on hyperscalers, etc., then we’re in a good place. But there will be a tipping point.

Dominic Bowen: Yeah, I think that’s really important and I think AI’s killer strength really is making the existing workforce much more efficient, not just automating tasks, but actually increasing efficiency. And I think I’ve heard others talk about AI as this kind of corporate Ozempic where it lets companies grow their revenue without the usual calories of hiring lots of new staff. And I think I need to give Professor Scott Galloway credit for coming up with that analogy. So I wonder, is that where we should be seeing the growth? Because we haven’t seen that. We’ve seen some companies blame mass layoffs because of AI, but not necessarily increase productivity. Is that the same thing or how should we be looking at that?

Craig Unsworth: Yeah, I think, again, it does come down to the size of company. In the mid-market space, I’m not seeing those layoffs and redundancies being attributed to AI artificially. I’m genuinely seeing processes, teams, functions being augmented with new tooling, which is allowing people to do more with less. And we started with a brief of do more with what you have. It is now very much in the do more with less, which means headcount reduction as well. I’ve got a few different examples from across the portfolio of clients I work with. One of them has taken a 15% headcount cut because of just adding new tooling. They’ve reduced all of the manual effort. It’s a legal tech business and that’s what they’re doing. That’s what they’re doing for customers. So they’re kind of drinking the Kool-Aid internally as well. I’ve also seen other companies say, well, we would have hired 35 new people. We haven’t. We’ve maintained our base at 150 heads. But we should have hired more. And we haven’t. We’ve put agents in. We’ve built tooling and changed our workflows instead. I think that’s more tangible. It’s more viable. It’s more believable. And it’s more visible. You can see it happening. I think if I were running a business of 100,000 heads and quite fancy getting rid of 15,000 of them just now, like we’re seeing with the large, large tech, AI would be a really good excuse. I don’t know how much of it is wholly authentic. I think it is happening. But I think that we are also seeing a natural leaning-out of business models.

Dominic Bowen: So what do you look for? What’s a serious CEO saying on their earnings call today to convince investors that their AI spend is actually disciplined and focused rather than just this empire building?

Craig Unsworth: I think it’s worth jumping in at the very beginning of that because I still judge quite harshly the investors who were ever letting vague spending go. They got caught up in the hype cycle and they should not have done that. So there is a change. I’m seeing more of a change in investor attitude than I am in company operations, actually. Most of the companies that I see at that level are still spending strategically, they’re still building that tactical defence, and they’re still trying to build and deepen a moat and defensibility. But the investors have changed their mind on what is vague, what is not vague, and where their comfort level sits. What I’m seeing just now is people shifting from just spending on AI experimentation and shifting from innovation into more operationalization and more productization of AI. So this time two years ago, people were saying, we’ll experiment with generative AI and see if we can replace some people. Now it’s very much: no, we’re building an agentic workflow, we’re replatforming our operational team, and we’re going to have 20 cluster agents and we’re investing in agent number 12 just now. It has a very defined spec, a very defined brief. We know what success looks like, and this is what the ROI has to be for us to succeed and move on to agent 13 in that workflow. I’m seeing much more of that. And I think that’s being driven by the narrative externally, especially with large tech, that there is bubble-like behaviour. I don’t think it is a bubble, but I think it’s bubble-like behaviour in some areas. You’re seeing individual engineers be hired with half a billion, a billion-dollar package is insane.

Dominic Bowen: And I think, you know, we spoke about NVIDIA before, but, you know, they keep reporting this extraordinary growth and, you know, talk about this enterprise adoption of AI agents is skyrocketing. But it’s certainly taken a hit from the Biden and Trump back and forth on the export controls tied to China. What does that tell us about the next phase of AI? Obviously, it’s no longer just a pure technological story, but there’s this geopolitical element in this competition. You know, what does that tell us about what we should be or what we could be seeing between some of the big players in Europe, America and China over the next year or so?

Craig Unsworth: I think there are a few interesting parts of that. So one is really how it has changed recently. And I think a lot of that is legislation-driven. We’re seeing the US, the UK, the EU and Asia all go in quite dramatically different directions when it comes to regulation, whether it’s how to extend GDPR for an AI world, whether it’s looking at individual data privacy, whether it’s thinking about AI defence, etc. And I mean, defence of the structures of AI not used for defence. But then we get to the world situation that we’re in just now, where defence is also with a capital D, front of mind. So I think there’s a lot of things going on just now. One of the things I’m really curious to see is whether or not we hold our nerve and continue developing AI with the original vision, which for some people, if you take the Sam Altmans of the world, etc., is this general intelligence position. But historically, we’ve been really bad as humans of getting halfway down a road with a new technology and then using it for something else. We’ve militarised it, we’ve weaponised it socially, and we’ve turned it into something that is about increasing wealth and revenue for a very specific niche. We’re quite bad at that. I wrote a Substack article about this recently, which I quite plainly titled ‘You Humans Might Not Be Smart Enough for AGI’. And I do think we have a track record of getting halfway down an innovation cycle and then completely messing it up because we act like humans. It will be very interesting to see if we act a little bit more like machines and what that does directionally here.

Dominic Bowen: And so how can individuals future-proof their skills and become someone who knows how to leverage AI, not just someone who gets replaced by it? I mean, is a mindset of this continual adaptation going to be enough or do we need to do more?

Craig Unsworth: I think it is enough just now. I think Joe Bloggs on the street can really develop enough of a skill set here with a little investment. One of the things I get involved with is mentoring and I’ve been developing a large number of relationships with underrepresented groups recently because I’m determined to try to play a part in this new revolution, not being something that widens inequality even further. And one of the things I’ve been able to do is work with groups of people who usually wouldn’t think that they had access to things here. And what we’ve been able to do is put together a really basic curriculum which is an hour a week and I’ve been helping supply issues, tasks and challenges with training notes. It’s incredible how quickly people in that group have been able to grasp the principles, the fundamentals and then some really quite advanced things. I’ve got people who, three or four months ago, had never logged on to ChatGPT who are now building agentic workflows in n8n. The workflows are really basic: they help with life admin or with school homework, et cetera, for their children, but it’s keeping them current. And I think investing an hour or two a week, like you would if you were trying to learn a language or get good at a sport or any other hobby, I think is a really important message to get through just now. I’d like to see that message coming from government and regulators about the individual responsibility we have to stay ahead of this. And that’s for everyone. Before you get into how else do you develop your T-shaped profile at work? How do you get career-specific tools? Because if you are sitting there in the legal profession, the old slightly clichéd adage now of, you know, an AI won’t take your job, but someone using AI will. That completely applies. There’s also the chance that AI might take your job as well. But in terms of getting there to a first line of defence, pick up some tools, experiment and start learning about the future.

Dominic Bowen: Public perception of AI really is centred around the large language models, Copilot, Grok, ChatGPT, Claude. But in your experience, what are some of the other AI tools that businesses, business leaders and young professionals really are overlooking and should be learning more about?

Craig Unsworth: I think you’re right. I think the LLMs are grabbing the headlines. Even things people call LLMs are not really LLMs, so people bundle Perplexity in there, which, of course, is not an LLM in its own right, but it’s doing a similar job. For me, the glue that connects everything together has to be your workflow. And right now, you know, I run an advisory business working with multiple clients, all of whom are private-equity firms or their assets. And I’m able to punch way above my weight on what I could have done years ago because I’ve leaned into agentic workflows. Little things like note takers, having every single meeting that I’m in, the note taker is there, making sure that I’m capturing every single action. I can look back and be absolutely sure what was said. I know how to bring quotes into documents, etc. All these things that would be so time-consuming otherwise. Diary booking, capacity planning. I optimize my diary every week, looking ahead and thinking, well, how do I almost like defragment my hard drive? And I kind of defragment my calendar and I use AI to assist on that. And these are all tools that are there. Gemini is incredible at helping me with calendar management. Claude and ChatGPT are great at helping me write up thoughts and notes, et cetera. I love playing them off against each other well and saying, well, ChatGPT gave me this summary of these notes. Claude, what do you think? And getting a second opinion. And suddenly, before you know it, I’ve got a virtual EA. I’ve got a virtual researcher. I’ve got a virtual archivist. I’ve got a virtual secretary, in the classic sense, writing up every single minute of every single call and meeting I have. All of these things come together to make a workflow that really ends up providing me with three or four FTEs—that full-time equivalent equivalent we discussed earlier—which, if I had to hire them, would make a rather bloated team for the work I do.

Dominic Bowen: Yeah, that’s interesting. And if we look at the company level, I mean, I think the majority of companies have started to at least experiment and integrate AI into much of their operations. But it’s a small number that are really doing this at scale. And I think that fear of missing out is quite significant. But with the companies that you’re working with and maybe the companies that you’re not working with as well, you know, how do they sensibly balance the caution required to avoid wasting money versus the risk of falling behind by not investing in AI now?

Craig Unsworth: I think the ones who are making the most headway just now got in early. They really did get in early. And I wrote an article recently about the impact and the cost of waiting with AI. It’s no longer a case of missing one thing and falling one step behind. It’s really, really changed. And that compounding effect of missing an opportunity is something that I’ve seen a lot of in the companies that I both work with and just observe in the business in the segment generally. The idea of getting started seems to cripple everybody. Where do you begin? How do you do it? And there are several people out there now that you would be able to reach out to and bring in as an AI enabler in your business. People who know your sector, who are technology-leaning and are able to actually assemble experiments and hypotheses, etc. So you can get started. I don’t think we’re going to have such a delay in adoption here. But I think the flip side of that is we’re going to have things, people, businesses, processes, products that just die because they’ve been completely replaced by something that did lean in and did embrace this new technology sooner.

Dominic Bowen: And speaking about leaning in, I wonder about where we should be concerned when we think about the risks. I mean, we talked before about labour issues and, you know, this isn’t a distant issue. I mean, Amazon confirmed 16,000 corporate job cuts in January of this year. Since about October, I think it’s about 30,000. Reuters has reported that this was blamed on artificial intelligence, shifting corporate culture. The January cuts were framed around reducing layers of bureaucracy and boosting efficiency. But at the same time, we’re seeing companies like Allianz, Dow and many others using what are now being termed AI-washing justifications. So, when we’re seeing major firms cutting thousands of roles while citing AI automation and AI-enabled productivity, should there be concerns about mass unemployment? And then, of course, looking at the second- and third-order effects, should our politicians and business leaders also be worried about what will no doubt follow, if that’s the case, around the social and political risks of this?

Craig Unsworth: Yes, a big yes to everything you just said. I’ve been looking at universal basic income for about 10 years and wondering what that could look like. The reason that you and I are sitting here at 3.30 on a Tuesday and not tending the field is because an industrial revolution happened. And that was net positive, it allowed us to be out there doing other things. This time, the net positive is far less clear. And I think we will have mass employment change. I’m not quite saying mass unemployment yet, but I think it’s definitely realistic that every single person’s job will fundamentally change in some way. What that looks like, we need to map out and examine. And if I were in a treasury role or a revenue role, I’d be looking at what that does to my tax profile. Because we’ve always assumed that AI would come for the low-paying jobs, with those workers being lower-paying tax contributors, so the impact being lower. At the point where you’re taking 30% of your legal workforce out, that has a different impact on the tax take at that point. We need to be modelling that out. And I hope somewhere, someone much smarter than I am and much more connected to central government is looking at this and absolutely modelling out scenario A versus B versus C, because I hear very, very little about how we fund this. What does this open up? What does this enable? What could the positive impact be? But, specifically, how do we make sure this isn’t a driver of poverty? How do we make sure it doesn’t eradicate the middle class, for example? These are all things that would have a significant political, societal and individual impact. There are going to be a large number of people in what they have always thought were very safe professions. You go to university, you get your undergrad, you get your master’s at business school, you go into consulting, eventually you make a partner, you retire, you play golf. That’s been defined and uninterrupted as a career path until now. You can replicate that across investment banking, you can replicate it across the legal sector and many other previously safe professions.

Dominic Bowen: Yeah, very interesting. I think it’s Justin Wolfers. He’s an Australian-born economist that’s at the University of Michigan. And I’ve heard him speak a few times, and he recently spoke about this case study. And I hope I do it justice. Otherwise, no doubt Justin will reach out. But it was that if someone came to you, Craig, and said, look, I’ve got this AI bot that can do your job as well or better than you, and it’ll only cost you $10 a week. You still get your salary from your employer. Your employer doesn’t care. Your employer just wants to make sure that you produce the same outputs every week. You can now buy this bot from me, and you get the same outputs. You get to spend the week down at the beach and in Majorca drinking piña coladas. Most of us would go, yeah, I’ll buy that. And I still get my same salary from my boss. That’s an awesome deal. Now that’s a great deal. And I think everyone goes, that’s fantastic. The company still benefits, but it probably does even better because now they don’t have to deal with my attitude. They don’t have to deal with me taking sick leave and they’re still getting the same output. But then Justin says, what about a different scenario where the salesperson, instead of going to you, Craig Unsworth, and asking, ‘Do you want to buy this?’ They go to your employer and they say, you employ 100 people. How about I sell you 100 of these bots? And instead of having to deal with Craig’s attitude and Craig being sick and you know the ups and downs of dealing with personnel, you’ve got these chatbots that will never give you an issue and will produce the same outputs for a tenth of the price. Now we’re all unemployed. Now the company’s got increased profits and that one person who owned that company has increased their profitability massively and reduced the difficulty of dealing with humans and we’re all unemployed. And he uses that as an example to talk about the extremely different way that AI could be employed and sold and utilised. One has a fantastic outcome for all of us, and the other has a really, really bad outcome for all of us. Have you considered those sort of scenarios and how do we manage that environment?

Craig Unsworth: So I love binary, black-and-white situations. My brain would really like to stick with just those two options. And you can navigate between the two, you can build a preparation plan for them both. The reality is I think it will end up being somewhere grey in the middle. And I think what we probably will have to ask ourselves at one point is how do we change taxation? So what would the normalised profitability be? What does an abnormal or highly varied profit look like with the example you mentioned of everything moving to agentic and having 100% unemployment? The profitability is changing significantly there. How is taxation shifting from income tax, which in your example has been lost entirely, to the equivalent of a corporation tax, for example? That’s going to be a really uncomfortable conversation to have because universal tax bandings, tax gradings, tax policy, etc. It is a one-size-fits-all policy. And what you describe and my rebuttal to it would require an individual, company-by-company tax policy, which basically just got every single politician zero votes. So I think it’s a really complicated time to be thinking about how you balance technology policy, how you look at fiscal planning, how you look at society and welfare. All of that needs to come together not only within one party, governing structure, country or bloc of countries, but globally. And I don’t know if you’ve noticed, but we don’t seem to be able to agree on anything globally just now at all. So I think that’s a real, real pending challenge for us is how we do that. Because first mover disadvantage would apply, you know, if the UK suddenly changes tax policy to be harder-hitting on AI profitability, for example, guess what? Every single AI startup, every single AI business will just decamp to Ireland or a different country that hasn’t enacted that policy yet. So I think we have a lot to do there. I’m not sure how positive or optimistic I feel about the people who are in charge in a space to regulate and legislate, but I would definitely like to see a few more of them leaning in a bit more heavily here with some experts to guide them as well. And those experts can’t be the same seven men who are all going to be trillionaires. That can’t be the entire expert panel.

Dominic Bowen: Unquestionably not. Especially when you hear these soft whispers about their concerns about the future of AI and then on the earnings calls or when they’re speaking to investors, it’s a very different message. Are there any risks in there that we’re just not seeing or that you’re not hearing the broader community talk about?

Craig Unsworth: Loads of risks. I guess we should talk risks because that is the subject here. There are so many risks. I looked at my risk register the other day. The average risk register I work on has gone from about 40 to 50 lines to around about 200 lines in the last 18 months. The risks I’m looking at primarily just now are moat and defensibility: how replicable is your business without a significant proprietary dataset, which doesn’t mean all of your data needs to be proprietary, but you need to process or aggregate it in a proprietary way. Your entire business could be replicated overnight. I used to sit in board meetings and investment committees and talk about moats being 12, 18, 24 months. I don’t know many companies that have a moat of more than 12, 18, 24 weeks just now. So I think that there’s a lot to be done there. And the other risk is, of course, that very, very well-tooled and very well-equipped humans in your employee base will be the ones who are A, most critical to you as a business, you want to keep them the longest; but, B, they’ll be the ones that could very easily leave and go and do something else as well. So I think we’re treading a bit of a talent tightrope moving forward as well. And then I guess the third one that I think about a lot just now is just the change in accessibility as well. So the opportunities that used to exist for very small companies or only for very large companies. How much of that is being rewritten just now? How quickly can you legitimately compete with a business that is now a thousand times your size? I think there are going to be examples where it’s completely possible, and that creates opportunity. But there are also going to be new examples where you have no chance. You might have had a tough journey competing in a David-versus-Goliath situation before; now, the game is over. So I think there’s a large risk there, before we even consider hallucination and inaccuracy. The issue people may be most reluctant to discuss is addiction. After nearly 20 years of social media, we’re finally beginning an adult conversation about social-media addiction. I think AI-assistant addiction and dependency, and the blurring of the line between machine and human, are going to become a societal challenge for us also.

Dominic Bowen: Thanks for unpacking that, Craig. And one question that we ask all our guests on the International Risk Podcast is when you’re looking around the world, and it doesn’t have to be within the business sector or within artificial intelligence, but when you do look around the world, what are the international risks that concern you the most?

Craig Unsworth: Defence comes to mind first. I think humans making decisions in a defence setting have led to some pretty disastrous outcomes in the past. There is hope and potential for machine-aided or machine-controlled decisioning to produce better outcomes, which could mean fewer people dying, which would be an incredibly positive thing. There, of course, is the risk that a bad human plus a bad machine can create an exponentially worse outcome as well. So I think that there’s a real issue there. And protectionism and trade protectionism, I think, are also significant issues. Who is going to fund the chip shortage? Who is going to fuel the water shortage? Who is going to take away resources from humans to give to machines? There are a lot of really big questions coming. The data centres that we talked about, we haven’t taken an overlay of where they are. But when you do take an overlay of where they are, and then you look at that mapping of where flood and drought risk is, etc. They’re all in horribly inconvenient places where actually flood risk is really high or drought risk is really high. Electricity consumption, you know, we’re nowhere near getting on top of clean energy yet as a whole world. I can be semi-smug, I suppose, coming from Scotland where we produce more energy than we use. The global market doesn’t reflect that in prices, availability, infrastructure, reliability, et cetera. Add in 20 data centres, you know, multiple football-pitch-sized data centres, and that antagonizes that situation. So I think that there’s a lot of stuff coming. And again, I can’t remember a time where we’ve had so many risks that would all come back down into one very narrow area of governance, either within a political setting or an enterprise setting. The job of the CIO in some of these large enterprises that we’re talking about just now is 10, 20 or 30 times what it was just five, 10 years ago in terms of responsibility.

Dominic Bowen: Yeah, it definitely is. And I think that’s an important point because we just assume that the CIO we had five years ago or 10 years ago and the executive teams and the skills and capabilities that they had are the same ones we need in 2026 when the operating environment is completely different. So I think that’s a really, really great point, Craig. Thank you very much for your time. And thank you very much for coming on the International Risk Podcast today, Craig.

Craig Unsworth: Thank you. Great to be here. And I look forward to seeing all the rest of your content on YouTube. I hadn’t seen your YouTube. I looked earlier before I came on, it’s an incredible wealth of content. So I look forward to joining that and exploring the rest of it as well.

Dominic Bowen: Fantastic. Thanks very much, Craig. Well, that was a great conversation with Craig Unsworth. I really appreciate hearing his thoughts on artificial intelligence and of course the potential risks and the opportunities. Please do have a look at the show notes below and we’ll link to some of Craig’s articles and his Substack. I’m Dominic Bowen, host of the International Risk Podcast. Thanks very much for listening. We’ll speak again in the next couple of days.