top of page
logo

AI, augmentation or Armageddon

Ben Graham-Nellor
Jul 21
6 min read

Updated: Sep 7

Twelve months ago the fear was mass unemployment. AI was going to take all of our jobs. It hasn't happened. And that fear seems to have gone quiet.


What's happening instead is more interesting. AI isn't replacing us. It's raising what's expected of us.


So, are we facing Augmentation or Armageddon? I'm still on the fence. My gut says augmentation, but it is far from certain.


I've noticed a change in my behaviour recently. I used to come home from work, do the usual house things, dinner etc, grab a glass of wine and sit down to watch some TV show I've been working my way through. I still do that. Except now, I usually have a conversation going with at least one large language model at the same time. I could be building an HTML questionnaire, updating the website, researching something I want to write about the next day or business planning by pushing ideas back and forth.


Although I have been using AI for years, this behaviour has been creeping up over the last twelve months or so. AI has become an integral part of my everyday life.


Why? Through my very tiny initial dive into the world of AI, I've come to realise that we're in the midst of one of history's great changes, an actual revolution that is going to change things, one way or the other, forever. I want to know a little bit about it.


This amazing change could go two ways, and I know the one I would prefer, but I'm still on the fence about which will win out in the end.


The Augmentation reality


This is what's happening in my own business right now. I've increased how many clients I expect myself and my team to serve. Up 30 to 50 percent, to somewhere between 150 and 200 clients per adviser. How? It's all in the efficiencies created by AI. Copilot has become another workhorse in our business, and we only scratch the surface of what’s possible.


Is that relief, or is that pressure? Exciting and scary at the same time. And when you're working this closely with something (someone? not sure how we address AI) you don't fully understand yet, it's probably both.


The optimistic story about AI at work goes like this: it takes the grunt work off your desk, so you get to spend more of your day doing the parts of the job you actually trained for. Client conversations instead of paperwork. Strategy instead of admin. That is my experience so far, and I’m enjoying it.


I believe in the augmentation story. I've lived a version of it. But there's a second half to it that needs to be thought through. When a tool makes you twice as capable, the expectation on you doesn't stay flat. It rises to meet the new capability. Nobody handed me permission to work less because AI is doing more. I just quietly started expecting more of myself. My guess is I'm not the only adviser this has happened to, and I doubt advice is the only industry where it's happening.


So augmentation isn't a free lunch. It's a raise in the bar, and we're the ones who have to clear it.


The Armageddon case


Now the other side. Unemployment and financial fear first, because that's the obvious place to start.


If AI genuinely displaces work faster than new work appears to replace it, the cost isn't abstract. It's someone's mortgage. It's someone's ability to feed their kids. It's someone's sense of who they are, because for most of us work isn't just a pay check, it's identity. What happens if that is taken away by an AI bot? More money for those at the top of the pile, but what about the everyday worker?


The crime story: How dignity in work keeps our society held together. It's not that people become criminals. It's that desperation does what desperation always does. Take away someone's income and their dignity in the same stroke, and some people will do things they never thought they'd do. Not because they're bad. Because they're scared, and desperate.


A Bocconi University study published in Econometrica, using Brazilian labour data, found that workers who lost their job in a mass layoff were 23 percent more likely to face criminal prosecution. The effect was sharper for economically motivated crimes, up 43 percent, than violent crime, up 17 percent, and it hit younger workers hardest.

That's not a reason to fear the unemployed (we all may soon be one of them, touch wood). It's a reason to take the financial side of this seriously before it becomes a crisis instead of a conversation.


But there's a gentler version of the Armageddon story too, and it's worth considering.

What if paid work being replaced doesn't have to mean ruin? What if some of that freed-up time goes to the garden you never had time for, the painting you gave up at nineteen, the music you always meant to write? For some people, mass unemployment might look less like Armageddon and more like permission. Permission to follow a passion.


Is that enough? Passion doesn't pay the mortgage or the rent. Which brings us to the harder question that society needs to consider.


Do we need a universal basic income?


A universal basic income, or UBI, is where everybody is given an amount to live on that is above the poverty line (unlike JobSeeker), there's no means testing, everyone gets it, even old mate Gina. If people then choose to do paid work, they are taxed. If they choose to live on the UBI and follow their passions, that's great too, and yes, some people will do nothing, but I believe that is a small percentage of people and shouldn't stop consideration of this idea.


How is this paid for, you ask? Company tax, dare I say a resources tax? If, in this Armageddon situation, wealth is flowing to the few, the companies who run the AI, then they need to be taxed appropriately to redistribute funds to everybody else.


This has been tested before, and it's worked well. GiveDirectly's universal basic income study in Kenya offers the clearest evidence yet that UBI can work. Running since 2018, it's the largest and longest-running UBI trial in the world, giving roughly 23,000 people across 195 villages one of three guaranteed-income arrangements, a 12-year monthly payment, a 2-year monthly payment, or a single lump sum of around $500, with 100 additional villages serving as a control group.


Full results are due out to the 12-year mark. However, so far, recipients show sustained improvements in income, savings, and self-employment, with no rise in spending on alcohol or tobacco, undercutting a common criticism of cash transfer programs. The standout finding: both the 12-year guaranteed income and the lump sum outperformed the short-term, two-year monthly payments on almost every measure, starting more businesses and earning more from them.


It turns out how you deliver the money matters as much as how much you give.


Sounds like a dream, doesn't it? It may well be the way out of a nightmare. The left hates it because there's no means testing. The right hates it because it smells like welfare. But if we do see mass unemployment at scale, I think it's worth a look.


I don't know if we're ready for that kind of restructuring. I'm not sure any country is. The hullaballoo after the last federal budget shows that any change is excruciatingly hard, but I'd rather we start asking the question now than when it’s too late.


Where I land


My preference is augmentation. My fear is Armageddon. I hope there are people in positions of power, much smarter than me, thinking these issues through properly. Because I don't think the rest of us can solve this one at a kitchen table.


The jobs that don't exist yet


"Won't somebody think of the children?"


Where do the jobs of the future come from, and how do we prepare our kids for them?

The honest answer is we can't. Those jobs don't exist yet. What we can do is teach kids to think critically, and to have original ideas. The technical skills, whatever they turn out to be, will likely be available to anyone with the right tool. The thinking won't be.

That's going to stay rare, so teach your kids to think deep and wide and they will be okay.


In the meantime, if you're not taking the time to learn about AI and where it fits in your world, I think you'll be left behind, and soon.

Whether that's a good thing or a bad thing is entirely up to you to decide.


One last thing


I can't write a post about AI without saying this. If you're using AI, you should be offsetting the impact it has on the environment. It burns a lot of energy. It uses a lot of water. And it's genuinely easy to offset, through groups like Greenfleet, who now have a product built specifically for this called AI ImpactCover.

It's not a huge commitment. It will make a huge difference to our world.


Go well.


Ben Graham-Nellor is a financial adviser and founder of Smart Happy Money, based in Melbourne. He writes about building a more human, more inclusive financial advice industry.


References

 

 
 
bottom of page