The AI Industry Is Starting To Unravel

The wheels are falling off the bandwagon.

Photo by Pawel Czerwinski on Unsplash

AI will revolutionise everything. It will be the next mega industry. It will beckon in a new, automated industrial revolution. It will separate growth from labour and generate copious amounts of wealth. This is the narrative that has been shoved down our throats by drug-addled techbros and the pathetic financebros that prop them up. This narrative has been used to leverage so much debt and investment into AI that we have basically bet the entire Western economy on its success. But there is one problem: this narrative simply isn’t true, and the AI industry is beginning to find that out the hard way.

Take Microsoft. Since 2020, it has invested multiple hundreds of billions of dollars into AI infrastructure and OpenAI. Its premier product from this colossal investment is Copilot, an ‘agentic’ AI to help you complete tasks in Windows, kind of like a jumped-up version of Clippy (god, I miss Clippy; it was awful in a good way).

Yet, it turns out no one is buying Copilot.

The Information recently reported that Microsoft has slashed its sales targets for Copilot after struggling to find buyers, with some targets cut by up to 50%! This is reflected in Microsoft’s financials, as they aren’t making much money from their huge AI campaign at all.

Microsoft did push back on the Information’s reporting, claiming they conflated sales and growth quotas. But that is like saying, “It isn’t horse shit on my face — it’s bull shit”. No matter what label you slap on it, it’s still horrifically bad to scale back either target! It indicates that the product they have spent hundreds of billions of dollars on is a dud!

But why is Copilot’s growth and/or sales so bad? Well, one study tested these ‘agentic’ AIs, including Copilot, and found that they flat out failed to complete even simple tasks 70% of the time, rendering them somewhere between useless and an active hindrance.

The same is true for Microsoft’s cash baby, ChatGPT. Like Copilot, it is often functionally useless. Some studies have found that newer models are ‘hallucinating’ at rates ranging from 33% to 79% for simple tasks. Even the very latest ChatGPT-5, which is insanely computationally heavy and blindingly expensive to operate, has a 10% hallucination rate for basic tasks. That is enough to render it pointless, as the time you will spend fact-checking the machine is more time than the machine will save you.

So, is it a surprise that investigations found that just 5% of ChatGPT users pay for the service? Not only that, but OpenAI has to undercharge these customers, meaning they are losing a significant amount of money for every paying user! And yes, that includes the $200-per-month plans. This complete lack of customer revenue is one of the significant reasons OpenAI is set to post an annual loss of tens of billions of dollars for 2025.

And this cautionary tale of zero customer demand isn’t just limited to Microsoft and OpenAI; it exists across the entire generative AI market. But guess what? It is set to get even worse.

With all this talk of hyperscaling and hypergrowth being thrown around, you might think that even if there are vanishingly few AI customers, this customer base would at least be growing.

But no, AI usage and spending are actually going down, not up, at least in the corporate world.

The Economist recently reported that the number of Americans using AI to “produce goods and services” was 11% in October. That is a decline from 12% of their previous survey. That might seem small, or within the margin of statistical error, but it is supported by other trends in this data. The same survey showed that, back in March, 74.1% of companies with 100 to 249 employees said they did not use AI, but that has now increased to 81.4%! For large companies of over 250 employees, that figure now sits at 68.6%, up from 62.4% in February.

On top of this, corporate AI initiatives are being scrapped at record rates. In 2024, just 17% were scrapped, but that figure has now jumped to 42%!

None of this should be that surprising. After all, MIT’s report, which found that 95% of AI pilots failed to deliver meaningful results, is now famous. This report has since been backed up by other surveys, such as the one by BCG, which found that only 5% of companies that deployed AI saw value from AI, or the one by Forrester Research, which found that 15% of their corporate survey correspondents reported an increase in profit margins from AI over the past year.

And this AI downturn is set to get worse next year.

After conducting that survey, Forrester Research analysed responses and predicted that in 2026, companies will delay about 25% of their planned AI spending by a year. In other words, because AI is so damn useless, more than a quarter of the corporate money previously predicted to be spent on it in 2026 will be held back!

This is made even worse when you find out that the majority of AI revenue for the likes of OpenAI comes from corporate customers, not private ones. After all, this is where AI is supposed to make its huge economic gains, so corporate AI spending should be going through the roof if that were true! But it isn’t true. Corporations are witnessing how flawed and useless this technology is and are starting to turn their back on it.

When you remember that a colossal stream of debt is currently the only thing propping up the entire AI bubble (read more here), this is horrific news. The AI industry has set itself up so that it desperately needs corporate AI spending to explode as soon as possible, just to cover the debts it is raising, let alone reach financial sustainability. As such, this economy-wide change in corporate sentiment towards AI is the canary in the coal mine. It is a warning that the propaganda that empowered this bubble is failing as it hits reality. It is a sign that the financial booby trap Big Tech has set for itself is about to go off in its face.

---

Thanks for reading! Don’t forget to check out my YouTube channel for more from me, or Subscribe. Oh, and don’t forget to hit the share button below to get the word out!

---

Sources: Futurism, The Economist, Extreme Tech, The Information, Futurism, Reuters, Motley Fool, CMU, The Register, Tech Radar, Mashable, BCG, Reuters, Forrester, Will Lockett, Will Lockett, Will Lockett