Why AI won’t destroy jobs
2025-06-01 10:10:38 +01:00 by Mark Smith
David Friedberg lays out a case against AI doomerism in the latest All-In podcast [05:34]:
“If I’m deploying capital, let’s say I’m a CEO of a company, and I can now have software that’s written by AI. Does that mean I’m going to fire 80% of my software engineers? Basically it means one software engineer can output, call it 20, 50 times as much software as they previously could by using that software generation tool. So the return on the invested capital, the money I’m spending to pay the salary of that software engineer is now much much higher. I am getting much much more out of that person because of the unlocking of the productivity, because of the AI tool, that I previously could. So when you have a higher ROI on deployed capital do you deploy more or less capital?
Suddenly you have this opportunity to make 20x on your money vs 2x on your money. If you have a chance to make 20x on your capital, you are going to deploy a lot more capital. And this is the story of technology going back to the first invention of the first technology of the caveman.
When we have this ability to create leverage, humans have a tendency to do more, and invest more not less, and I think that’s what’s about to happen […] people assumed […] you can create a whole startup with one person, you don’t need to have venture capital anymore to create a startup. In fact what we’re about to see is much more capital flowing into tech startups, much more capital being deployed, because the return on the invested capital is so so so much higher, because of AI.
So generally speaking the premise that AI destroys jobs is wrong because it doesn’t take into account the significantly higher return on invested capital, which means more capital is going to be deployed, which means far more jobs are going to be created, and far more work is going to get done.”
First of all, I found the earnestness of his critique of the folks exagerating AI fears by he himself using a since-the-cavemen analogy, kind of funny. But putting aside wild exagerations for a second, I do think he makes a very interesting point, which could at least be, to use another All-in-ism, directionally correct. It makes a lot of sense. But I think it’s worth remembering that not all people in the marketplace will be thinking this way. The trick if you are a software developer will be to find people that are thinking this way, and of course to be fully immersing yourself in AI development tools.
Anecdotally, yesterday evening while listening to a Bitcoin podcast I had an idea for an app and decided to try and vibe code it on my phone while listening to the pod. Suprisingly I had something working within a few minutes. It was on the surface very impressive, but when I dugg a bit deeper there was a load of backend stuff that was conveniently mocked out. There would definitely need to be a lot more work to get something operational and production ready. But it did get me from 0 to something within a few seconds, and sometimes that’s half the battle. It’s definitely worth trying to use some of these tools in your daily work, just to get a sense for where they could speed things up. It might not be in the places you expect. They definitely are not a drop in replacement for humans in my experience. #