There Is An AI Bubble
There Is An AI Bubble
An economic jack-in-the-box is about to go off.
Anyone who had the misfortune of being self-aware during 2008 is painfully cognizant of the concept of an economic bubble. For those lucky enough not to be, or for those who have dissociative amnesia with 2008, an economic bubble is when shares or assets soar in value way past what they are really worth, thanks to unfounded speculation. This inflated “bubble” will then “burst” when a market correction unveils the genuine value of these assets, spooking investors and buyers, leading to mass sell-offs, and, therefore, crashing the valuation of assets, which leaves many high and dry. While 2008 was a spectacularly big bubble-bursting event, even smaller ones, like the dot com bubble, can decimate entire industries. With this in mind, is the AI industry, with its grand promises, insane stock valuations and investor hype, in a bubble? I think so. Not only that, but it is ripe to burst in the next few years. Here’s why.
Let’s start with a problem I’ve covered before. AI is being misused.
AI has proven time and time again that it isn’t reliable enough to replace humans, and won’t be for a long time. Instead, it should be used to create tools that augment and enhance human workers. This way, you get AI’s benefits without any drawbacks. This advice hasn’t come from an ethics committee, but from the Harvard Business Review, which has a track record of knowing a thing or two about how businesses should run themselves. To top this of, an extensive MIT review found that AI is currently way to expensive to replace most jobs. Yet, if you listen to any leading AI company or startup, they are positioning their products as being able to disrupt and replace human workers. This positioning is not only unfounded and separated from the reality of AI as it is today, but it falsely inflates their share price, and investors speculate they could acquire vast amounts of revenue by directly replacing workers.
This is where the bubble comes from. AI technology itself is incredibly powerful, but this miselling and overvaluation is creating an economic bubble. Don’t believe me? Let’s look at two examples.
Tesla is one of the most well-known AI companies, with its self-driving AI making it one of the highest-valued car companies of all time, thanks to investors speculating that their self-driving vehicles could replace taxi and truck drivers and push sales way up. That isn’t hyperbole, as Ark Invest has predicted that Tesla could be worth 11 times what it is today by 2027, thanks to its AI. However, Tesla’s self-driving systems are still miles away from being fully self-driving, and they are facing a mountain of federal investigations and lawsuits over their dangerous nature. Not only that, but as Tesla uses a computer vision-only approach, many in the industry believe Tesla can never reach full self-driving (read more here).
In 2023, Tesla made a gross profit of $13.4 billion, and its current valuation is $535.98 billion. But Toyota, which has technology that could eclipse Tesla, including its own gigacasting, AI self-driving systems (though marketed as a driver assist system), and affordable super-long-range solid-state batteries, and a gross profit of $61.699 billion, is only valued at $331.30 billion. These numbers are the smoking gun of an AI bubble.
OpenAI is another example of this value false narrative.
Their generative AIs like ChatGPT and Sora have been hailed as industry crushers, able to wipe out an entire section of our economy. In their current guise, they definitely can’t. ChatGPT, for example, continues to hallucinate false facts and fumbles basic grammar, and Sora still creates videos firmly in the uncanny valley. However, OpenAI and investors have stated that the subsequent iterations will solve these issues by using larger training datasets, making them way more capable. This, in turn, has pushed up the speculated value of OpenAI.
Yet, OpenAI CEO Sam Altman has warned that these larger models will use so much energy to develop, train and maintain that they will require an energy breakthrough like nuclear fusion to be viable. You see, even with the most affordable energy we currently have, the energy bill for these next-gen AIs will render companies like OpenAI unprofitable.
This is another example of a bubble. Investors, market analysts and alike are valuing OpenAI way above where it should be, even though its CEO has admitted they have a baked-in upper limit to what it can do. Funny enough, these same energy cost limits also apply to Tesla.
If that doesn’t convince you, then how about the recent poll by Bank of America that found 40% of fund managers think that AI-related stocks are in a bubble.
So, the question must be: what will the fallout be when this bubble eventually bursts? Well, I’m no financial analyst, and even financial analysts have difficulty answering questions like that. It depends on which funds or banks have bet the farm on AI and who has correctly mitigated this potential risk. But if/when this bubble bursts, it isn’t likely to take out the leading players. Tesla, for example, is incredibly cash-rich, and its business isn’t entirely based on AI. They could survive even the worst AI market crash if they don’t waste all their money on dead-end AI development and keep their EV sales buoyant. It would be severely kneecapped, but it won’t die. The same applies to other AI giants like Nvidia, Google and Meta. Sadly, businesses solely based on AI, or AI startups heavily leveraged to grow, will almost certainly not survive such a burst.
I genuinely believe an AI bubble burst is on the horizon, and judging by how badly shares like Tesla’s are doing right now, I also think the market correction is only around the corner. I really hope I’m wrong, as this could severely financially damage a vast number of people. I guess only time will tell.
Thanks for reading! Content like this doesn’t happen without your support. So, if you want to see more like this, don’t forget to Subscribe and follow me on BlueSky or X and help get the word out by hitting the share button below.
Sources: Bloomberg, Bazinga, City AM, Euronews, HBR, Investopedia