The deluge of speculative tech investments unleashed by artificial intelligence is unparalleled in history. Never before has a technology which promises so much — but currently does so little — managed to capture enough funding to threaten the US economy should it fail.
With that kind of cash on the line, one would assume that tech startups have done their homework on the complicated reality of AI development before courting untold millions of dollars from overzealous investors — that’s sort of their job, after all. But according to Facebook’s former head of design turned tech entrepreneur Julie Zhuo, the hottest AI companies on the market are getting by on “good instincts and good vibes” alone.
In a recent interview with podcast host Lenny Rachitsky that was spotted by Business Insider, the tech industry veteran laid out the remarkable reality of the recent explosion in AI investment: tech companies are growing fast, with no interest in understanding why.
“I don’t think a lot of the fast-growing companies are using data well at this point,” Zhuo explained. “And, the main reason why is because, traditionally, things just don’t grow that fast.”
Zhuo outlines what could become an grim irony if the near-infinite well of venture capital ever starts to dry up, a major anxiety for companies like OpenAI. Essentially, she tells Rachitsky, AI startups that got by on vibes and hype alone will find themselves screwed when it comes time to function like an actual business — precisely because of their explosive growth.
“Today we see companies that are growing insane,” the former Facebook exec continued, “and they’re still about ten people, or two people, or however many people, but they’ve got hundreds of millions [in revenue], and hundreds of millions of users, and they don’t actually have all of that infrastructure… to be able to do that data analysis.”
Of course, it’s not a question of “if” the music stops; experts have been hollering for months that the current AI funding boom is financially unsustainable.
“In my mind, [data] helps us reflect back on what is reality… what always happens is eventually, things stop growing,” Zhuo cautions. “Growth does not happen forever. And usually when growth stops, everyone has this question like, ‘what’s going on, what happened?'”
Taking data and analytics seriously, Zhuo explains, makes it much more likely that a company will diagnose the root cause of slowing growth, or even that a company can predict the inevitable financial slowdown in the first place. “If you don’t have good observability over how your business runs, then you will be scrambling,” she said.
Zhuo isn’t necessarily concerned with the broader economic fallout of AI hype. As a businessperson, she’s more interested in value from the point of view of a startup’s overall longevity.
That said, she doesn’t have to be: some 33 US-based AI startups have raised over $100 million in 2025 alone. A handful of those funding rounds were in the billions. So far, none of them are even close to turning a profit on AI, something even the leading big tech giants can’t seem to figure out.
When the cash inevitably stops flowing, the cautious, data-driven AI startups may be the only ones left standing — if any of them survive at all, that is.
More on AI hype: Data Shows That AI Use Is Now Declining at Large Companies