Reasons I think the AI bubble will burst this year; Why my dad's doctors failed to diagnose his illness
1) As I've been saying for years, I strongly believe that artificial intelligence ("AI") is revolutionary.
That feeling has only been reinforced since Google's Gemini quickly and correctly helped diagnose my dad's recent illness after doctors had failed for a week, as I shared on Friday and yesterday. (I'll also discuss this further at the end of this e-mail.)
And AI keeps improving at a rapid rate. Shopify (SHOP) CEO Tobi Lütke recently posted on social platform X:
I shipped more code in the last 3 weeks than the decade before. The top AI models / agentic systems right now are an entirely different thing to what people used until the beginning of December.
That said, I think there's a clear spending bubble in the AI sector that's likely to burst this year. Today, I'd like to outline the reasons I believe this. And tomorrow I'll discuss the investment implications.
The company I'm most worried about, which I think will be the trigger for the bursting of the bubble, is the privately held maker of ChatGPT, OpenAI.
It's incinerating cash at a rate I've never witnessed before, yet its model has fallen behind Gemini. I think it's going to run out of cash in the not-too-distant future, the effects of which will ripple through the AI economy.
2) Regarding Gemini's superiority over ChatGPT and why this gap will likely expand, this Business Insider article highlights how Alphabet's (GOOGL) "Google is flexing its advantage over AI rivals like OpenAI and Anthropic" with its ecosystem of apps that have billions of daily users.
This X post from tech entrepreneur and investor David Lieb shows how Gemini's integration with other Alphabet services helped him schedule a haircut:
Meanwhile, he shows that ChatGPT failed in this same task:
Tech influencer and AI consultant Ashutosh Shrivastava concludes (correctly, in my opinion):
3) As for OpenAI running out of cash, let's start with the fact that it's losing money at a staggering pace. It's projected to have lost $8 billion last year, rising to $14 billion this year.
Its stock-based compensation – albeit a noncash expense – is out of control, even by the extreme standards of the tech sector, as Charlie Bilello points out in this X post:
Journalist and author Sebastian Mallaby thinks OpenAI will run out of money for reasons he outlines (and I agree with) in this New York Times op-ed:
Generative A.I. businesses are not like the software successes of the past generation. They are far more capital-intensive. And while behemoths such as Google, Microsoft and Meta earn so much from legacy businesses that they can afford to spend hundreds of billions collectively as they build A.I., free-standing developers such as OpenAI are in a different position. My bet is that over the next 18 months, OpenAI runs out of money...
However rich the eventual A.I. prize, the capital markets seem unlikely to deliver.
The probable result is that OpenAI will be absorbed by Microsoft, Amazon or another cash-rich behemoth. OpenAI's investors would take a hit. Chipmakers and data center builders that signed deals with Mr. Altman would scramble for new customers. Social media pundits would report every detail, and frazzled investors may dump the whole A.I. sector. But an OpenAI failure wouldn't be an indictment of A.I. It would be merely the end of the most hype-driven builder of it.
Investor George Noble shared similar views in a recent X post. Here's the first part:
4) There's also the theory that emerging companies squandering money on expensive Super Bowl ads are heading for trouble. This Wall Street Journal article notes just how much OpenAI could be spending:
OpenAI is expected to air a Super Bowl ad for the second straight year, escalating its battle for users as rival AI companies gain ground.
The artificial-intelligence company behind the popular ChatGPT tool is set to broadcast a 60-second commercial during NBC's broadcast of Super Bowl LX, according to people familiar with the matter...
Some advertisers are paying more than $8 million for just 30 seconds of airtime during this year's Super Bowl, according to ad buyers. Those prices don't include the millions or tens of millions that brands often spend to craft new ads for the big game. It couldn't be learned how much OpenAI would be paying for its spot this year.
This will likely just add to the company's growing pile of lost cash.
5) Lastly, here are a few anecdotes from my readers...
Goran H. writes:
I would echo the praise of Gemini. I have canceled my ChatGPT subscription and now use almost exclusively Gemini. I think it's deeper, more specific, and just more helpful overall.
Steven L. adds:
Gave Gemini a try this morning. Entered the same question in ChatGPT as well as Microsoft Copilot, and Gemini gave more information than the other two. I'll use it more often. Thanks for the tip.
In tomorrow's e-mail, I'll share my thoughts on the investment implications of a bursting AI bubble. In particular, I'll note which stocks to hold onto and which to avoid. So stay tuned!
In the meantime, here at Stansberry Research, we're sharing our "Game Plan 2026"...
This is our complete investing strategy for the year, including a buy call that could double your portfolio. And the best part is, there's less risk – even if we see the AI bubble bursting or a similar market sell-off.
Click here to get all the details in this free broadcast.
6) While it was Google Gemini that helped correctly diagnose my dad's illness, I'm having great fun taking full credit and telling all my friends and family to please address me as "Dr. Tilson" from now on!
The most common question was: "How on earth did the doctors not immediately figure out that your dad had African tick bite fever? It seems like they were totally incompetent."
In part, the answer is bad luck: The nasty lesion from the tick bite (which you can see a photo of in Friday's e-mail) was on the back of his leg, so they didn't notice it initially.
But it was a total failure to have not done a careful examination of his entire body when they first admitted him. Worse yet, they didn't even ask the most basic questions that might have tipped them off, such as: "Have you done anything out of the ordinary in the past week or two?"
He would have answered: "Why yes, I got back nine days ago from a 10-day holiday in South Africa."
This would have led to a follow-up question: "Were you only in cities, or also out in the bush?"
Answer: "Five of the days, I was doing one- to three-hour hikes in the bush."
Seeing the lesion and hearing this answer surely would have led the doctors to quickly focus on an illness resulting from an insect bite, right?
Not so fast... While they initially didn't see the bite or ask about my dad's recent travels, my mom said a day or two after he was hospitalized, she flagged both of these with the doctors, and they still didn't react!
How is this possible, given that these are the best doctors in Kenya who are very experienced and were trained in the UK?
The answer is that they were all cardiologists. You see, my father has had atrial fibrillation for a number of years, so all of his doctors are heart specialists.
My dad's main symptom was extreme weakness, so they were running all sorts of tests related to his heart. They fell into the classic trap captured by the saying, "To a man with a hammer, everything looks like a nail."
Ironically, if my dad had shown up at a random hospital where he didn't already have a team of doctors who knew him, I suspect he would have been seen by a generalist doctor. They would have done a body scan and asked the right questions, identified his illness within 15 minutes, put him on doxycycline, and he would have been healed and home within a day or two.
There's an important lesson to be learned here, not just related to doctors: Anytime you're dealing with a specialist in any field – health, investing, law, etc. – you need to both respect their expertise, but also recognize the blind spots and often-perverse incentives they can have.
For example, if you have back pain, a surgeon is generally going to recommend surgery, a physical therapist will extoll the virtues of physical therapy, an acupuncturist will want to stick needles in you, etc. Warren Buffett and Charlie Munger have long talked about this – one of Buffett's favorite sayings is, "Don't ask a barber if you need a haircut."
I'll let my good friend, Tom M., a doctor who runs a major hospital system, have the last word:
Your dad's case is fascinating from a U.S. lens, but I have to say this is bread-and-butter basic health care for folks abroad. How on earth did they miss this obvious parasitic infection?
His eschar with the dark patch is so classic – it's unmistakable. All I can think is they didn't do a very basic full physical exam, including undressing him and looking for a rash all over his entire body. A first-year medical student could have caught this if they looked.
But mistakes happen, and AI is helpful for sure in these situations, especially for lay folks trying to solve a mystery, so great job.
We use it now, and it is being embedded in our electronic medical records and processes, but with oversight always! AI is not evidence, but it's a good source and readily available to help out.
Best regards,
Whitney
P.S. I welcome your feedback – send me an e-mail by clicking here.