Veteran investors knew this reality check of extreme valuations was coming sometime.
It would be easy to throw in the towel on artificial intelligence (AI) stocks. Not only have they (as a group) gone nowhere since October 2025, several of the technology industry’s stalwarts, including Microsoft and Oracle, have been struggling since reporting last quarter’s earnings. Investors are starting to wonder if these companies will ever be able to justify the massive investments they’re making in AI.
And this concern isn’t without its merits. A survey performed by PwC earlier this year indicates that over half the world’s corporate CEOs haven’t yet seen any measurable benefit to utilizing Ai, while a study done by MIT suggests that, as of last year, 95% of generative AI projects didn’t provide any net return on the investment. Not good.
Image source: Getty Images.
But it certainly wouldn’t be the first time a new business idea ultimately didn’t live up to the hype. Think meal kits, or non-fungible tokens (NFTs).
Artificial intelligence stocks aren’t doomed, however. That’s because AI has true enduring value. The crowd just needs to adjust its lofty expectations. These stocks will struggle in the meantime while they’re rethinking what these tickers are actually worth now, and what they can plausibly be worth in the foreseeable future.
Going through the Gartner Hype Cycle
Veteran investors have seen it before, of course. The phenomenon is so predictable, in fact, that it even has a name. It’s called the Gartner Hype Cycle, so called because it was recognized and given specific, named stages by technology consulting outfit Gartner. See if you can identify where artificial intelligence is in this five-stage adoption cycle of a new technology.
Innovation trigger: Something that wasn’t possible before has become possible.
Peak of inflated expectations: The market’s imagination about what this new tech could do and how it could reward investors runs wild…
Trough of disillusionment: …until reality sets in. As it turns out, just because a new technology can achieve something doesn’t mean that something has value or is profitable.
Slope of enlightenment: As time marches on, new practical and marketable uses of this tech are identified and developed, while less fruitful uses fade away.
Plateau of productivity: Finally equipped with a full understanding of what this technology is and isn’t good for, companies only commercialize products with an obvious potential for a sustainable return on the investment.
And several nascent modern-era industries have worked their way through Gartner’s hype cycle shortly after their beginning, including solar power, 3D printing, autonomous vehicles, and virtual reality, just to name a few. All of these technologies are in use today because they have real value to mankind. But the key stocks of all of these industries also ended up punishing investors who insisted on getting into a growth opportunity as early as possible, but were expecting too much too soon.
The dot-com craze of the late 1990s and the subsequent crash of the early 2000s, of course, remains the most obvious and egregious example of the Gartner Hype Cycle. Companies like Amazon and Microsoft were already publicly traded back then, when access to the internet became common — the innovation trigger — and both stocks are certainly priced much higher now. But both stocks were also shellacked in the midst of that meltdown during the trough of disillusionment. They didn’t really begin their long-term rallies until the slope of enlightenment took shape in the early 2000s and led into the plateau of productivity thereafter, when the crash’s survivors finally turned profitable.
You’ve got a little time to figure out long-term winners
You probably already figured out the AI industry is currently in the midst of a trough of disillusionment — everyone agrees it’s amazing, but not everyone sees the actual practical benefit of its offerings right now, particularly at their steep cost. Investors are now starting to seriously scrutinize the returns on these investments so far, and aren’t liking what they’re seeing. That’s why powerhouse stocks like Microsoft and Oracle are trading down more than 20% and more than 50% (respectively) from their AI-inspired peaks reached just late last year.
These two big names are, of course, weighing on shares of their peers.
The challenging part here for true long-term investors is accepting that this disillusionment will need to run its course until we reach the beginning of the slope of enlightenment — the point at which artificial intelligence provides enough real purpose at a price that developers can offer such tools profitably. That could take a while, however. It might take well into the middle part of the year, in fact, preventing these tickers from making forward progress in the meantime.
That turnaround is coming, though, sooner or later, and probably sooner rather than later. Your big job between now and then is just figuring out which AI names are the Amazons and the Alphabets of the artificial intelligence revolution, which is undoubtedly here to stay.