Job Security: Optimised for Our Own Obsolescence?

Gareth King (00:52)

Nick, thanks so much for joining us and welcome to Ruined by the Internet? To start us off, when it comes to work, the promise of tech has seemingly always been to liberate us from not only the workplace but the hours as well. In your opinion, do you think that this promise has been delivered, or have we devolved somewhat into an always-on work culture enabled ironically by that same technology? Like, what's your take on that?

Nick Jain (01:11)

Yeah, I think you're absolutely right that the promise of technology has been more free time for recreation or consumption. And what has happened in practice is people end up working more. And I think there's a very fundamental reason for this. When you are able to work more, that means your competitors are too, your colleagues, the guys you went to school with, the guys who are working in the cubicle next to you. And someone there will what's called, it's like a classical prisoner's dilemma.

Even if you all are better off if no one decides to work more because technology is doing more of your work, somebody will decide to work a little bit more. Since they can now work from home, they'll decide to work 12 hours a day instead of eight. And you, to keep up, have to do that. And it's a race to the bottom that is enabled by technology but is really it's not technology's fault. It is human psychology's fault that we are competitive.

Someone out there is uncompete, is out competing you, and you have to keep up. And technology just enables, you know, call it some of the worst of our natural behaviours or inclinations.

Gareth King (02:06)

Yeah, that's a great point around that competition and there's always that competitor that will not take the foot off the gas, so to speak. But we know a lot of talk around technology in general, especially in the workplace, does, as you’ve just mentioned there, relate to productivity.

But do you think with such a high velocity of new tools to constantly be across, especially right now, do you think that pressure to constantly be retraining and upskilling across everything or the all the new developments that are constantly emerging into the market, do you think that ends up playing a role in what's often mentioned as declining productivity amongst especially digital based workforces?

Nick Jain (02:43)

So yes and no, there's something called the J curve. Are you familiar with that at all? So okay, so yeah, it's like whenever you start something new, you lose a little bit of productivity, you go down and then you go up. And the whole point is when you're engaging in a new activity, and it could be a new skill, if you're learning how to run a different way or throw a ball a different way or use a piece of technology, you have to stick with it long enough so you get out of the bottom part of the J curve and start rising up. So, it's actually worth it.

The problem is if you switch technologies or procedures or protocols or companies too often, you're always stuck in that bottom part of the J curve. So, it is okay to be switching technologies quickly, but for any given individual or any given organization, you need you need to understand how long you're going to be spending down in the J curve.

Gareth King (03:25)

Just on that on that point there around the J curve, you know, we can kind of just get into that from that continuation now, the AI that we know is huge right now and there's a lot of talk around AI, whether it's scepticism, optimism, fear, you know, bullishness, whatever it is, or a mix of everything.

One thing that we often see the most AI forward people claim is how they've 10xed their output or whatever it is, simply by utilizing these tools themselves. Now, on a surface level, what's the argument that that that 10x boost to what they're delivering in the same amount of time, for the same amount of wages if you're a wage earner, is not just a ninety percent devaluation of their previous labour? Like how do we square that circle?

Nick Jain (04:13)

I think it assumes, the kind of the negative view is that you assume a fixed pie. So, if everyone is 10x more productive, it means that everyone can work ninety percent less and therefore you get ninety percent less wages. But that assumes the consumption of productivity doesn't increase. And I'll give two good examples from history.

About six thousand years ago, when all of our ancestors were running around the fields and forests hunting and gathering, we, you know, didn't have a lot of food. And then someone or a group of people figured out, hey, we can farm, we can plant forests. And all of a sudden, our productivity in producing food went up roughly somewhere between 100 to 500x.

Did that mean everyone just laid around, lounging around? No. When you had more food, what happened? More uses for food came up. More people came into existence. So, we rapidly consumed more of that resource. We found ways to grow and evolve our society. And that's the more optimistic view that AI is going to result in a such a productivity boom that we will find new ways to apply all that productivity that we've unlocked that just we, it didn't make sense to before, right?

6,000 years ago, it didn't make sense to have cities because you would starve and die, until you had that hundred X improvement in food productivity. We saw the same with the Industrial Revolution when we moved from steam power to electric power. Electric power was way, way more efficient than you sticking your turbines next to a river somewhere. And we're probably going to see something very similar with AI technologies that we are going to find more interesting ways to consume it that didn't make economic sense three years ago.

Gareth King (05:45)

Is there a ceiling to how far that that, I guess, new utilization of that productivity can go, especially if it's not something as fundamental to us as people as, say, food production?

Nick Jain (05:57)

Well, you say, you know, food is a fundamental thing and I agree with that. But I think a lot of people would dis would say that there's lots of other fundamental things. And I let's use a silly example of video games, right? Video games are such a core part of human life that they are now, you know, beginning to rival pro sports teams in terms of franchise value.

That technology or that entire lifestyle, that resource of video game consumption and all the ecosystem that exists around it, was only enabled by about 30 or 40 years of increases in computing power.

So, what you are kind of what you or I might call unessential is in fact incredibly essential to lots of people. So, there's lots of cool things that are going to arise in the post-AI world that you or I may say, hey, this is not part of what it is really to be human. It doesn't involve eating. It doesn't involve relationships. It doesn't involve procreating or making babies. And but others will say, well, this is so cool. It is important to my way of life.

Gareth King (06:51)

Look, that that's a fantastic point there around I guess the creation of new ways to utilize that. You know, and the example you gave there was the compute power when it came to video games. Now, I guess my question from there is does something like that rely on someone having that interest in video games and the creativity to imagine new ways of utilizing it?

Because I think where I'm going with this is the topic of displacement when it comes to AI specifically right now. So, if we are eventually ending up with so many displaced white-collar workers from whatever section or industry that they're from, will people that are potentially displaced through the utilization of AI initially, are they usually the ones that will find new ways of utilizing this, this additional productivity or ability?

Or is it still gonna come down to say the people that can leverage it the most, uncovering new ways to bring those ex-employees back into different roles? Obviously not in the same company, but just in a general sense. Like where does this go, do you think?

Nick Jain (07:58)

You know, earlier you asked me if I was an optimist or pessimist. I'm an optimist on what these technologies enable. Unfortunately, I'm a pessimist on what it means for the current workforce. So, I think there's a lot of people who will be displaced in a fundamental structural way. And we can draw the analogy to the American auto industry.

When automation started happening in the middle half of the 20th century, people said, okay, well, these guys can go retrain from knowing how to use a hammer to being machine mechanics. And that's all nice in theory, but in practice it doesn't happen. And a lot of those people just end up fundamentally displaced.

And I can, you know, just very bluntly, I can offer a practical example from what we're doing. So, we're building an, we've built an AI powered CFO and are out there selling it, right? Are we doing rocket science? No. But can the average person who spent a decade training in finance or in business build what we do? No, it requires a certain amount of aptitude in technology and coding and design that ninety percent of white-collar workers don't have.

Now, of that ninety percent, maybe a third can actually go learn these skills and reskill themselves, but that still leads to, you know, sixty percent of displaced workers. The more positive lining is this change is not going to happen over the next year. We're going to see the impacts of this rollout over a decade or two decades, not immediately the next year.

Gareth King (09:12)

Yeah, that's an interesting point that you've raised there around the timeline of it. It's being massively disruptive right now, and it feels like the fear is quite high and people are starting to realize like, hey, maybe I was a little bit too cynical of this thing. It seems to be progressing a little bit more than I gave it credit for.

Is that what's really happening on the ground at the moment? Do you think that people, as you've said, should be optimistic of what they'll potentially be able to do in the future? And then how does that lead to that example that you mentioned that you've implemented yourself?

Nick Jain (09:42)

Sure. So again, I think that these technologies are, the good news is they have very thin parts of the J curve. Almost anybody can pick up some of these AI technologies, not all of them, and be productive instantly. And as a personal example, I suck at all things graphical. Yet it is very easy for me to pick up Midjourney or Dall-E or I think Google's is nano banana 2. And within seconds generate an image that would have, you know, I would not just would have taken me hundreds of hours to make before, but I could have never made it that quality.

So, I think the J curve is actually much thinner for these technologies where it makes sense to adopt them. But there's a big difference between adopting these technologies and building them from scratch. And I think that's where a lot of people are going to struggle because you're going to have these people who build these technologies that all of a sudden displace a hundred workers or a thousand workers. And those thousand workers are not all capable of building the technologies for themselves that find the next new niche.

Gareth King (10:35)

That's a great summary. Not everybody is capable of being an entrepreneur. That's way daunting for so many people. And I think I think my concern on that point is what happens to those people? Now, I think it maybe it was like a decade ago or something that the phrase “learn to code” kind of entered the lexicon as a suggestion for what would have been traditionally blue-collar workers who saw their industries being automated or something else for various reasons. They were told learn to code.

Now, obviously, 10 years later, people who've studied coding and you know program whatever it is, i'm not an expert on that. They're at this wall now where so many of them are like, you know, I use Claude or I use this tool or whatever. I have only written one line of of code in X amount of months. I don't do anything. My job's being eradicated. I see it from the inside.

Now these feel like people that can utilize these tools. For someone that might have been told learn to code, may have like picked it up and now they're facing this eradication of their entire career path, and we see people being told, learn a trade now. You know, it does feel like we've come full circle. And I think that, yes, the tools as you've alluded to are super powerful if you gain the understanding of them. But whether everybody is capable of doing that, I don't know.

And I think where I'm going with this is there's like a real focus, and we can see it in layoffs that are, you know, being attributed to AI efficiency. Ten people, 10,000 people, whatever. All of it results in that same level of output or desired level of output, but with less people doing it.

And that brings me back to what you were saying you know a little while ago there, which was that productivity boom, is that still going to be the aim? Because that theoretically would rely on the same amount of people 10Xing their output, or will people be more likely or businesses be more likely to say, No, that's the level we need. We'll have one person doing it. Nine of those ten people, sorry, no job for you? Like what's happening to the masses here?

Nick Jain (12:34)

So, I think there's a slight difference in terms of whether you look at it at the individual level or at the organizational level. At the individual level, these technologies, even if you are non-technical and you just pick up ChatGPT, you can do or Gemini or pick your favourite model, you can all of a sudden do, you know, incredible things and not just chat things. You can build websites, you can make music, you can make videos, you can code, whatever.

At the individual level, it's a definitive yes, this increases your productivity. However, human beings don't do things individually. They do things in organizations, whether that be churches, schools, companies, governments, et cetera. And when you think about it at an organization level, an organization has kind of two or three things going on.

Firstly, they have the total productivity coming from the sum of their individual employees. Second is what they have what's called a coordination task. If you have 10 people, put together, they may be 10 times more productive. But if you put a hundred together, they're not going to be a hundred times more productive.

Because now we have to manage all those people that communicate. They have to spend time on Slack and email and just explain, hey, Gareth, I'm working on this, you know, across the hallway from you. That's called a coordination tax.

And the bigger you get, the more you pay. Imagine, you know, imagine you're a multinational company, you have to coordinate across languages, time zones, cultures, departments, business lines. And then the third is like, okay, how many ideas, how many business opportunities are there that we would love to be working on? And it's a conflation of those three factors.

And I think for some organizations, you're going to find that they just have a great business model. They know what they do. They don't have the next business line. For them, it's just for you know, productivity means they can half their workforce or cut their workforce by, you know, 80%. And we're so we saw that, for example, at Block, which is Jack Dorsey's firm that laid off 40% or plans to lay off 40% of their workforce due to AI technologies, or due in part to AI technologies.

And for them, the implicit the implicit statement is hey, we have no other ways to grow other than doing our core business. And our core business is doing so well, we can just do it with fewer people. There's other organizations where they will be like, okay, this productivity means we can go chase all those other ideas we just didn't have the time or money or resources to do before.

So, whether that's a layoff in or in terms of shrinking the organization or keeping the organization static, it really depends on how many incremental business opportunities there are for the organization to expand into. If there's not many, they will shrink their workforce. And you're, I think you're going to see that at a lot of companies, right? If you're a typical manufacturing company, the demand for your products is not driven by, you can't just invent new products for the most part. So, it's driven by external macroeconomic forces. And for those, you're they're going to be chopping more jobs on average than they'll be creating.

Now that may be very different in a software firm or a law firm or professional services firms.

Gareth King (15:18)

Yeah, that's a great point there around the ability to chase new opportunities. Where I'd love for us to go with that now is what companies or what type of businesses would you say are best poised right now to explore new opportunities through this efficiency or productivity gains or whatever it is through these tools? And which ones from your perspective are kind of worse positioned?

Nick Jain (15:43)

I'll make two categories. I think firms that are not making physical products, I think will on average have more opportunities to pursue than those making physical products. Physical products are really determined by how much physical stuff people need. So, I think you know, manufacturing firms or things are they're gonna focus much more on cost efficiencies than they will on we can make a new product.

There will be some exceptions. For example, AI has shown the ability to design new engineering materials better than human beings. So that may result in explosion of, although those are not manufacturing firms, they're research firms, right? They research it, they develop a new material or new vaccine, and then somebody else goes and makes it.

So, the but going back to that first bucket, for firms that are providing an intellectual service, these could be technology firms. So actually, I think that there's two big groups that are gonna be winners. Number one are software firms that are gonna make new types of software that it just didn't make sense to do before because it might cost you a million dollars to crank out a piece of software that now you can do in a few thousand dollars.

The second is entertainment firms, broadly defined, Hollywood, images, video games, music firms, because the cost of producing the entertainment is going to drop dramatically, meaning that the solo artist can now create a lot of music. Or conversely, a big Hollywood studio that only has a hundred-million-dollar budget, sounds like a lot, but the typical Hollywood movie now costs a hundred million dollars. So, they can make one big movie a year.

If their productivity doubles, they may actually they're these are creative people for the most part, they would love to go make ten movies. They just didn't have the budget to do it before. Ten X more efficient. They can go make those ten movies.

Gareth King (17:13)

Yeah, look, this there's two points there that I'd love to get into with you now. The first one was around cost efficiencies. And just while we've just looked at it, that creative industry, you know, 10xing output, whatever it is. Before we get into the cost efficiency thing, like let's just focus on the ability to make so much more output, say if you are a Hollywood studio.

And it seems that the general perception largely from what I can see is they are big, they've got money, they don't need to be using this stuff. So, it almost creates a backlash in amongst the audiences. And that is before we get even into let's say you're an actor, whoever you are, you're the most famous one, you're commanding, let's say, eighty million dollars for a film.

Will they A still be able to do that? And then what's the flow on effect once you can produce a film for so much cheaper, and their likeness is potentially up for grabs now? You know, if we do have these studios producing, I don't know, a hundred films for the cost of one previous film. Does that just flood everything with so much stuff that it's almost overwhelming for people?

Where do you think that that Hollywood example or that related industries plays out before we get into just the focus on cost efficiencies?

Nick Jain (18:27)

Hollywood's a great example because we're thinking about Hollywood the way it is today, where you have actors who, you know, command giant salaries, actors or actresses.

But if you roll back the clock on Hollywood specifically, American Hollywood, go back about a hundred years. Actors were not extra, they were famous, but they were not rich because what happened was, they're they were developed like products by a studio where a studio would hire a young actor, begin promoting him or her, but then they would basically be locked into 10 or 20-year contracts that with that one studio. So, they could not just go work for another studio.

So, what that meant was who owned the actor? The actor was basically an asset or an owner, not a free agent of Hollywood. They couldn't just go negotiate with the studio next door and say, Warner Brothers is not paying me enough for this movie. How about you pay me something a bigger number for that movie?

Now what happened somewhere in the I think don't quote me on this, but I think it's somewhere in the sixties or seventies, there was a specific event. But basically, actors became free agents. They were able to pick whatever movie they want. They had their own independent representation. At that point, they became their own products. And the studios were their customers.

In today's world, sticking with actors, you're right. We're going through a legal thing about who owns the actor's likeness and voice and stuff. Let's imagine that lands in the actor's favour, that they are still their own asset or their own brand. That's great for them. They'll continue to monetize it.

But what I think you're going to discover is a lot of Hollywood studios are going to make up celebrities from scratch. So rather than having it look like Brad Pitt or look like Gareth King, it's just going to be a made-up human being that they can have in multiple movies, a persona.

And a fun example of this that you see from again going back to video game culture is the Master Sergeant from Halo, which by the way I've never played. But it's a franchise, right? A guy in a mask wearing a certain uniform costume or Doom Guy from a game that I've you know, I played thirty years ago. These are franchises where, you know, there's not an actor, there's not a person behind the mask, or even if there's a person behind the mask, it's just a made up character, and yet some company owns the intellectual property behind this name and likeness, not a human being, because it wasn't based on a human being.

I saw an article about there was the first successful fake influencer.

Gareth King (20:45)

Yes I was gonna bring this up in a minute. I'm glad that you have. This is an interesting case study.

Nick Jain (20:49)

I didn’t double click too much into it because, but you know, it's not surprising that there's an influencer out there who just doesn't exist and is able to sell whatever products she, it ,is selling and it works.

Gareth King (21:01)

Yeah. And that that it's just on that that case there, it is so interesting because it does seem like the majority of people are really not that bothered. Do you know what I mean? Like they're not thinking about it as this is not a real person. They're just taking it as the communication tool that it is. So, it will be very interesting to see how that all plays out.

But I'd love to get back onto the point around cost efficiencies that you mentioned a few minutes ago. Now, one of the debates that I've seen quite a lot is around this notion of AI washing, which, you know, a lot of people are saying that this rush to appear at the cutting edge and implementing AI due to a struggling business model is essentially just AI washing - huge amounts of redundancies, or even offshoring or outsourcing work remotely, all being framed as optimization via AI.

Now two things there. If it's not being delivered by AI now, are they just offshoring and outsourcing it to say a country where the work can be produced for fifteen percent of the local value?

Nick Jain (22:06)

Can you define AI washing just 'cause I I've actually have not heard that term before.

Gareth King (22:10)

Okay, so AI washing would be saying, hey, we need to lay off twenty percent of our workforce because we're going to be implementing various AI tools, but the theory or the cynicism behind a claim like that is that the AI is not actually sophisticated or doesn't perform well enough to actually replace those people. It's actually just a euphemism or a code for, you know, offshoring work to much lower cost workers, which is obviously enabled by technology again.

Nick Jain (22:45)

Got it. So, I guess I have three responses to that. Number one is does it really matter if you're the worker who's laid off, do you really care if it's because some you know, AI's doing your work or someone in in a different country's doing the work or you know, if you if China's changed macroeconomic trade policy, it ultimately doesn't matter. You lost your job.

Gareth King (23:06)

No. No, sorry, just to want to interject on that. Yes, like totally agree. You've still lost your job. And I think I think the reason that from my perspective, why it is so contentious is AI washing is just throwing more fuel onto what is potentially a really hype driven fire that every other company looks at. It's like, shit, I've got to get on board this AI thing and implement it too. I've got to slash my own workforce. Like how much of it is hype, how much of it is hope, and what's the reality?

Nick Jain (23:36)

Look, I can AI do every job? No, right? I'm a mathematician by training. Can it do high school level math problems? Yes. No problem. Could it do some of my senior college level math problems? No. Will it probably get there in a matter of years? Yes. But today at least, it cannot handle complex multi-step logic problems. And you can that's true in finance as well. Basically, anything that involves depend somewhere, depending on how you measure it, somewhere between 10 to 100 steps, AIs tend to get lost.

Now that's still, now in all fairness, that's still smarter than 90% of human beings already, but it's not better than the best 10% of CFOs or lawyers or doctors or whatever. But guess what? The best 10% of doctors means there's still 90% of doctors that AI can beat, right? And so, it's not all hype. These technologies have really gotten to the point where they are more capable at mundane tasks that human beings do every day.

So, like a silly example is for a generic diagnosis, right? For like, hey, you've got a cold. An AI is going be better at figuring that out than a human being much more quickly as well. Now, if you've got some rare disease, sure, human being you need a human being or a doctor. But for some you know, traditional things, which by the way, are 90% of what most people do in their day-to-day job, AI can do that more effectively. So that's not hype.

Second, is look, at least in private sector and Western economies, there's an incredible pressure to get more profitable every year. So, Google, they may be at 35% profit this year, I'm making up the number, but there's still pressure from their investors, from their board. Can we get from thirty-five to forty percent? And that means the board is saying that to the CEO and the CEO is executing on that on his or her plans.

So, whether it be AI or hey, it's just cheaper in another country, there's going to be pressure economically saying, like, even if we're doing well, we need to do better. And if we're not doing so well, we definitely need to do better. Otherwise, we'll be out of business next year, right?

And by the way, that's something people often miss for companies that are not doing well. Their choice, the choice might be lay off a third of your workforce or lay off a hundred percent of your workforce a year later, right?

Survival you know, capitalism is ruthless. If you don't do it and your competitor out competes you, all of your employees are gone. And, you know, it it's truly that. Like if you get better, you can make your workforce better off, but you're gonna hurt everyone else. It's kind of a not a zero-sum game, but it's a, it hurts.

Gareth King (25:52)

It's that's a fantastic point. And I totally agree with you on that. Two things that you said there. Whatever the reason, you still lost your job, fine. Like that, it's not really going to affect anybody any differently. It's more the headline story of everyone needs to get an AI because it's replacing workers and then to your point then around the competition, it's like, my god, my competition is more ruthless than me. I've got to compete with them and get more ruthless, etcetera, etcetera. And so if we play that out…

Nick Jain (26:21)

It's not that they're necessarily more ruthless. They are they've discovered, you know, if your competition discovers the way that's that they can perform better, one of two things is going to happen. Either they're going to be more profitable, and they're going to buy you and then implemented that in your business. Or number two, they're going to cut prices, take your customers, and you're going to go out of business.

So, in capitalism, again, like if your competitors find a way to be better than you, only two things happen. And number one, they buy you because they're richer than you, or number two, they take all your customers away and you shut down. That is the unfortunate, ruthless fact about capitalism. And that is true regardless of whether you're talking about AI technologies or other technologies.

Gareth King (27:00)

Yes, absolutely. Now, obviously, if you gotta compete, you gotta do what you gotta do. As you said, capitalism is ruthless, business is gonna business. We you know, that's, you know, just the way it is, unfortunately.

But when you when you mentioned there around, you know, potentially if someone utilizes these tools, because a lot of the cost of many goods or services has got the cost of effort and labour baked into it. If we are reducing so much of that cost of effort and labour out of the production price, does that lead us into price deflation?

Or do you think that people will be able to maintain their historical, you know, effort and labour inclusive prices, even though they're very clearly using these new tools which remove so much of that?

Nick Jain (27:46)

So, in it, there's a good answer to that one, which is look, if you are in a monopoly where you're the only one selling your product, then you can maintain those higher margins, and you've captured all that productivity. If you're in any other industry where there's roughly three or more, there's an economic rule, I think, where you have if you have four more of four or more competitors, someone is going to cut prices when they discover a technological innovation.

And you see that in almost any other industry, right? Like a good example is phones, right. Yes, Apple is still out there selling an extremely expensive phone, but for everyone else, right, there's been a ruthless price competition. So, the what the phone that you're getting is incredibly, incredibly valuable relative to what they could sell it for.

If you look at every other phone that's made by Samsung, Google, whatever, yeah, they're they seem expensive, like five hundred dollars or whatever, but they're not actually, the actual profit margin that these companies are making is quite small because they are pushed by Motorola makes a phone, Samsung makes a phone, Huawei makes a phone, Google makes a phone, right? So, you get so much pressure that it actually benefits the consumers.

And you see that in almost any productivity gain in in any non-monopoly industry. And by the way, most industries are not monopolies, right? If you're, if you don't, if you're not one of the top ten tech companies in the world, you're basically not running a monopoly. You have to be price competitive. Boeing and Airbus, they're in an industry where there's only two companies in the world and they still have to compete pretty ruthlessly against each other. Notice none of them are making billion dollars of profit a year, right?

Gareth King (29:13)

Yeah, yeah, yeah. Look, that's a that's a great example that I can tie back to what you were saying earlier. Like the more opportunities will open up, the more competition, which could potentially make it better for consumers until…

Nick Jain (29:24)

Definitely gonna be better for consumers. Definitely better for consumers, not necessarily better for producers.

Gareth King (29:30)

Where does that leave us, you know, when you say producers, is the, I guess the second order effect of that less workers? You know what I mean? Like there'd be obviously a huge drive to cost optimise as far as possible on something like that.

Nick Jain (29:44)

So, look, there's going to be a lot of legacy players in services businesses or technology businesses that are going to go bankrupt because somebody can rebuild what they did much, much faster, much cheaper without technical debt or organizational debt. And those organizations are just going to, you know, go away entirely because they're going to be outcompeted.

The next set of organizations are those that are going to evolve, but they will tend to evolve more slowly. And so instead of having a $3 billion consulting firm, you're going to have a billion-dollar consulting firm because two-thirds of your business were taken away by emergent, by new competitors that are going to emerge from smaller organizations.

The net net of this, still, like, and I this is where I'm a little bit more pessimistic, is in an era where AI can be reasonably intelligent. And I don't mean artificial like generalized artificial intelligence, but rather where it can do work that historically required a human brain and decision-making judgment to do, I think a lot of those jobs are going to be permanently gone. And because AI is a generalist, it I'm not sure a lot of those displaced workers will find work in the new world.

And again, I'm not saying this with a, I'm saying this quite negatively. I think we are headed towards very much maybe not a dystopian society, but the only way that we will function is with some form of universal basic income. Whether or not we politically have achieved that is a separate issue.

Gareth King (31:05)

Yes, look, I mean that's a that's a great segue to UBI, but I think just on that, you know, work displacement stuff, one thing that we never got into is the your example of how you've implemented those tools yourself and from there, like what the outcomes have been.

Nick Jain (31:20)

Sure. So, I'll provide two examples. One is so we just launched a business. So, it's just me and my business partner right now. But here's the cool thing. Almost all of our work internally is done you using Claude Code. So, our emails, our marketing, our blogs, our website coding, the analyses we do for our client, and it is all done in Claude Code.

So, we're not even opening up, you know, I don't even open up my Gmail inbox anymore. I just have it, you know, an AI agent do it for me, summarize, rewrite the emails. I'm, you know, sometimes I look at them, sometimes I don't, and just hit send.

The second so that's internal productivity but then let's talk about what for a second about what my external product is. So, what we built was an AI CFO. And our entire hypothesis was look, if you're a billion-dollar company, you can hire a really smart CFO to help make your business better. But that person costs a million dollars a year. The average business can't afford a million dollars a year.

What if you're a, if you're a five million dollar a year business or $10 million a year business, you'd love to have somebody that smart being your CFO, but you can't afford him or her. So, our idea was there's all these people who would love to have CFOs, but they can't afford to have a world-class CFO. So, let's build a piece of technology for them.

So, if our hypothesis is right, then we have not taken away anybody's job, but we've created value for our or for our customers. And so far, it's working, right? These are all customers that didn't have CFOs in place. They would have loved to have a CFO, but they didn't want to spend half a million dollars on that. Now we're giving them a piece of technology that costs forty or fifty thousand dollars a year. Good, you know, it's good for us, but really, it's great for them because now that that piece of technology is helping them make, I think for the average client, they paid us $10,000, I think, over the last two months each. And on average they made about two million bucks.

Gareth King (33:10)

And the way that you've just broken it down there, I think that that is that soon to be revealed opportunities that a lot of people will be able to capitalize on and make. Now, obviously, what you've built there, there is a market for that, which is bringing that very, very high-level knowledge, ability, whatever it wraps up into those companies that can't generally afford to pay for that.

But I think my question on that when it comes to not taking anybody's job, yes, on paper it's just another service going out into the market. But is the reality that these companies would have had a, let's say a lower-level financial controller. But now that they've got your tool that comes along, which, you know, as you said, it's not directly taking anybody's job. It's providing a quality of ability, knowledge, everything like that, as you've just outlined.

But does it in a way replace that person? I think if we play that out to its logical conclusion, as these tools get smarter and that all that agentic AI gets better and better at replicating that human experience and knowledge, where does that at an aggregate play out once I guess the average worker is no longer needed? Do you know what I mean?

Nick Jain (34:21)

Yes. So let me break apart where we are today. So, our technology today does not replace a lower-level worker or even a different worker. So, if you think a CFO does a very different job than a controller, a controller is really focused on accounting, and a CFO is focused on business advisory. How do we make the business more profitable? Cutting costs, adding revenues, decreasing risk, whatever. One is an analysis job, and one is a record keeping job. Both are important, but in different ways.

So at least today our technology is not really replacing any accountants, controllers, treasurers, et cetera. That might change in the future, either with our technology or someone else's, but at least today, we're literally not replacing anyone. We're literally adding a function that they could not afford before.

Now, if we roll forward to the future, yes, you're really going into kind of that pessimistic prediction I laid out, which is these technologies, if not today, definitely in the next five or 10 years are going to be smarter than almost any white-collar digital worker. Not just smarter, but they can do your job. And that, it's a future that's coming reasonably quickly, I think well within the next decade. And the only limiting factor is how quickly organizations either adopt these technologies or are forced to adopt these technologies by either shareholders or competition.

And yeah, we look, again, I'm quite pessimistic on what this bodes for the white-collar workforce over the next decade. I'll use me as an example. If I wanted to spin up a world-class marketer, I think I can, somewhere between a w a day to a week is what it would take me to code one up, stick an AI brain in it, and all of a sudden, I have an employee that would cost me $80,000 a year in a week.

And after that, by the way, if you ran any agent 24-7, like 100% of the time at full capacity, roughly they cost about eighteen hundred dollars. That is cheaper than outsourcing to a lower cost country. And no human being works twenty-four seven three sixty-five. So, the technology is almost, either there or almost there.

Gareth King (36:19)

Yeah. Look, and that's a great point there. That if we and again, me just as a layperson, I'm trying to play that to its logical conclusion, where it just becomes so, such a no-brainer that why would you have humans working when you can be doing this? And then it makes me think where we started back off with the promise of that leisure time, not just much, much more productivity slammed into the same amount or even more work.

But one thing I've just remembered we didn't get into. I'd love to hear your thoughts on if we do get to that point where I guess the 24-7 nature, the very, very low cost of these agents displaces almost all white-collar workers, as you've mentioned. Regardless of whether we do get that leisure time back to indulge in our passions like a lot of people imagine we will be able to, or we end up in some kind of dystopian hellhole, the topic of UBI. It does come up a lot. A lot of people theorize about it. Me personally, I'm fairly cynical that it would ever actually come to fruition without some gigantic political shift.

I'd love to hear your take on it as an idea and whether it could happen or not.

Nick Jain (37:24)

So, I think that there's firstly I do genuinely think we're headed towards some sort of socioeconomic cataclysm where a lot of white-collar workforces are going to be displaced. And that's particularly difficult for Western economies that are largely knowledge economies, not labour not, you know, physical goods economies.

So oddly enough, AI may not really matter so much for China or India, but may absolutely devastate, you know, Australia or the United States or England or Canada or whatever.

So that's number one. So, when you end up in that world, and I think we're, you know, in our life, in well within kind of yours and my lifetime, Gareth, I think we end up in a situation where a democratic nation is going to vote in a UBI favourable political party. And it will be a fundamental shift away from the kind of laissez-faire capitalism that we've had for most of the last, I don't know, five hundred years in Western countries.

But it's going to happen because the alternative is, you know, there's going to be riots in the street. So, I do think UBI is going to happen. However, the thing people don't talk about is how much, right? Right. How much money is the government going to give you or our AI overlords going to give you? And I think it is just enough to make their consciences feel good, or keeping you happy enough that you produce whatever else they still need you for.

Kind of, and that's not a high very high number, right? The amount of money that you need to give me, let's use me as an example, to prevent me from picking up a gun and storming, you know, starting a riot or anarchy, it's not actually that high. Right. It's very easy to keep people happy with a very modest amount of income.

Gareth King (38:55)

No, absolutely. And I and you know, it's in the name, isn't it? Basic income. And one of the other side arguments that I've seen in discussions around UBI is, well, without that motivation to actually go out and earn, what's why would people do anything?

And I don't know, I find that a really weak argument. Simply because I think as humans, we are driven, we're curious beings and a curious species. But if those opportunities are still there to do more? Like you've got everyone's got their needs taken care of, that doesn't mean that everybody will settle for that. People will still want to maximize their ability. So, I think it is a very interesting concept. Like I said, I'm a little bit cynical.

Nick Jain (39:40)

jump in there for a sec a second? So, you said, look, not everyone will want to settle for that. You're right. But the key word you said is not everyone. I think if you go look at the economics research, lots of people are okay just being average, right? Or just getting a minimum amount of income. Most people are okay just working a nine to five job and earning $50,000, $60,000 a year and not wanting to go for that $100,000 job or that million-dollar job.

So even in a UBI world, I think you are right that some people will try and achieve more. But I think you'd be perhaps negatively surprised at how many people are just going to be, this is good enough, I will live and not die.

Gareth King (40:19)

No, I agree with you. I think a lot of people will take that approach to it and, you know, utilise that leisure time. I just think that the argument around, well, why would anybody do anything more? I think it doesn't take into account our… Yeah.

Nick Jain (40:33)

That's a terrible argument. Some people are just creative and driven, right?

Gareth King (40:37)

Totally. Yeah. Yeah. Yeah. Um yeah, it'll be interesting whether it actually happens or not through, like you said, the AI overlords being taxed for each displaced worker to pay for things like that. Yeah, so I guess we'll all be keeping quite a quite close eye on that one.

But you mentioned before that ten-year timeline. Is that your estimate of it? Or do you think it could be less? Do you think it could be longer?

Nick Jain (40:57)

Here's my kind of prognostications. Number one, I think the technology is already there to replace the majority of white-collar workers. Okay. So, I think the generally a reasoning LLM, so the top one or two models from each of the big AI companies is already smart enough to do better than the average, you know, employee in almost any white-collar function.

The reason I say 10 years is two things need to happen. Number one, organizations need to either agree to adopt these technologies or be driven out of business or outcompeted by people who do. And number two, even though the LLM is smarter, you still have to train it on doing a specific job.

And I don't mean training in the AI sense but rather, think of an LLM as a jet or an AI as a generally smart human being. You have to tell them, hey, this is your job, go do this. Just like you have a week or month-long onboarding with employee. So, you imagine you're a large company, you have you know a hundred thousand employees that do, you know, a thousand different jobs, you will then have to teach the AI, or onboard the AI to do those hundred different jobs, a thousand different jobs, right.

So that's why it's going to take 10 years. Not so not because the technology is not smart enough, but because you have to onboard the AI and tell it, hey, go do this job.

Gareth King (42:07)

With that, how do you think the future of jobs plays out from here in a general sense?

Nick Jain (42:12)

My two cents is that we're going to see a lot more independent founders show up, number one. And number two, we're going to see a mass displacement of white-collar workers over the next decade or so, about a third of which will adapt to the AI native world, and maybe two thirds of which are going to really struggle in the new world that we're entering. I'm not saying that is a good thing. I'm just saying genies out of the bottle. And I think that's the way we're headed.

Gareth King (42:40)

Well, Nick, you've definitely given us a lot of very interesting and some frankly quite daunting stuff to think about. Just to finish up then, what's a belief or concern that you have about technology right now that you wish was getting more attention?

Nick Jain (42:53)

So, I think everyone's talking about AI in terms of how smart it's getting, whatever the next new model is, whatever the next new app is. The thing I think people are not concentrating enough on is the cost of AI. And what's really, really dramatic is that the cost of AI or the cost of AI tokens is dropping by, depending on how you measure it, somewhere between seventy to ninety five percent a year.

Gareth King (43:18)

Nick, thank you so much for joining us today. What have you got coming up and where can people follow what you're up to?

Nick Jain (43:24)

For people who want to get in touch with me, I think the best way is to go to our company website, eaglerockCFO.com. There you have our email, our LinkedIn, as well as the opportunity to try our technology totally for free. Again, eaglerockCFO.com. Thank you so much.