AI Will Destroy The Economy
One way or another.
The notion that the current AI industry is in a negative-sum game isn’t new. So much money has been dumped into this technology that we have practically bet our economy on its success. So, if it doesn’t succeed, we become the newest residents of recession town. But if this bet does pay off, then millions will lose their jobs, and our economy will collapse, sending us to the same bleak place. It’s shooting ourselves in the foot whilst screaming “innovation!” However, we are starting to see the details of how these scenarios might play out. So, let me explain.
If AI Fails
If you have read any of my work before, you know I am slightly sceptical of AI, so let’s start with the AI failure scenario.
We are now starting to see how AI could fail.
Quite simply, it isn’t the productivity tool it was sold as. A Harvard Business Review study found that AI didn’t reduce work, but consistently intensified it. The employees “worked at a faster pace, took on a broader scope of tasks, and extended work into more hours of the day, often without being asked to do so.” In other words, AI isn’t a productivity tool, but a burnout tool. This is supported by the recent report from Oxford Economics (read more here), which found that companies “don’t appear to be replacing workers with AI on a significant scale” and instead suggest that they are actually using the AI layoff narrative to cover up their own shortcomings. Meanwhile, Forrester principal analyst JP Gownder has said that they simply don’t see any boost to productivity from AI in any dataset. Senior US economist Ronnie Walker found something similar, stating that “We still do not find a meaningful relationship between productivity and AI adoption at the economy-wide level.” The only areas he says had productivity gains were “customer support” and “software development tasks.” Note that it isn’t software development, but specifically granular tasks within the job. That doesn’t mean it equates to software development jobs increasing productivity, without decreasing quality, as there could be (and almost certainly are) bottlenecks that AI doesn’t address.
So, even though it might feel like AI makes you more productive on an individual level, that isn’t scaling to an economic-wide impact. Which is a shame because it needs to do that.
That isn’t just my opinion, but that of Nobel laureate Geoffrey Hinton, often called one of the “Godfathers of AI” for his enormous contributions to the artificial neural network technology that powers AI. In an interview with Bloomberg, he was asked whether the eye-watering investments in AI will ever pay off. Hinton replied, “I believe that it can’t,” and elaborated, “I believe that to make money, you’re going to have to replace human labour.” Other reports back this up. As I covered in a previous article, a recent report found that the AI industry will need to generate $2 trillion in annual revenue just to pay for the data centres they plan to build by 2030. This report used very optimistic revenue projections to estimate that the AI industry will be $800 billion short of breaking even by 2030!
To give you an idea of that scale, Google, one of the most monopolistic companies on the planet, made $400 billion in revenue in 2025. So, just to pay for the datacentres AI runs on, the industry will need to pull in more than four times that, and AI can only generate that kind of revenue by replacing huge swaths of the job market.
So, what happens if AI doesn’t replace labour?
Well, the entire industry will be cannibalised by its own debt.
