# Do You Have AI Psychosis? Turns Out a Lot of CEOs Do, According to the CEO of Box **By:** Joe Procopio **Published:** 2026-06-02T10:00:00+00:00 **Source:** [Inc](https://www.inc.com/joe-procopio/do-you-have-ai-psychosis-turns-out-a-lot-of-ceos-do-according-to-the-ceo-of-box/91353718) --- Finally! We’re finally talking about CEOs having “AI psychosis.” Over the last two weeks, I’ve been hammering the Corporate Tech industry’s penchant for loosely adopting AI via mandates thought the organization, running wild pet projects to push the AI-native narrative, firing large chunks of their tech workforce using AI as cover for poor decisions, and then wondering where all the ROI went.  I’m an AI OG, one of the inventors of some of the first commercially available generative AI, back in 2010. I’ve actually been writing about these implementation errors for a couple years now, on top of 16 years of being Don Quixote tilting at AI adoption windmills.  Now, I feel like Gene Hackman in Enemy of the State, the crazy guy who finally gets one of the heroes to listen to his mad ramblings. I feel, as the kids say, seen.  Because today, we’re finally talking about all this in the mainstream. And Box CEO Aaron Levie is calling it “AI psychosis.” It’s not, but whatever. He has a great point, and the writer of that article has many more great points, basically echoing the warnings I’ve been screaming from my digital bunker in the links above and dozens more. AI psychosis? Do you know what this means? I’m not crazy, they’re all crazy. But it’s not psychosis, and it’s certainly not crazy, and it doesn’t end here. We have an opportunity to correct years of unforced errors in the adoption cycle of AI.  Here’s what AI psychosis really is, how to know if you’ve contracted it, and how to beat it. Why We’re Finally Talking About AI Psychosis Thank you, Aaron Levie and Julie Bort. Levie called out the CEO behavior I describe above. Bort, a journalist, backed up everything Levie said with additional reporting and facts.  And people are finally listening.  I mean, I get it. I’m not a fancy billion-dollar-company CEO and I’m not a hard-hitting tech journalist. I’m just a fly in the ointment, a monkey in the wrench. I’m just a long-time techie entrepreneur trying to make things better for other techies. So as much as I hammer bad CEO decisions and the sometimes unquestioning coverage of those decisions, I know I also need to call out the leaders who lead with grace and the tech companies that aren’t pouring gasoline on a dumpster fire. In this case, I need to laud Levie, a tech CEO who says: “CEOs are uniquely prone to AI psychosis because they’re sufficiently distant from the last mile of work that still has to happen to generate most value with AI.” I need to give heaps of kudos to Bort, who adds: “Levie advises CEOs to use AI ‘a ton’ to really see what it can and can’t do, ‘and come out the other side with an appreciation for both the upside and the real work.’ I have enough faith in humanity to believe that there are CEOs out there attempting to do just that, but right now, they seem to be in the minority.” I’m in love. With both of them.  But is this really AI psychosis? Nah. That would imply that this is an unconscious strategy. That it’s something that happens to a leader, not something they enact as a means to well-thought-out ends. I have enough faith in (and experience working with) CEOs to understand that even the most maverick among them very rarely act without approval, let alone thoughtfulness.  I don’t want to make an epidemic out of the behavior of a few bad apples. What AI Psychosis Really Is If you don’t want to go down the rabbit hole of those links I posted in the opening of this post (you should, but you do you), AI psychosis is really just a vicious cycle of FOMO, mistakes made as a reaction to that fear, and then trying to fix those mistakes by making more mistakes.  I’ll TL;DR it for you.  The vicious cycle starts when leadership, usually at the behest of shareholders and their boards (but not always), get swept up by the fear of not understanding the depths of the newest AI trend, and are beholden to whatever the experts are telling them they need to do.  So the leaders subscribe to the AI-as-labor-replacement narrative based on what they do know, like how automation wrecked blue-collar jobs decades ago. They use correlation to get to a false causation, and start poorly implementing half-assed AI that mimics the other guy’s half-assed AI. Then they hit the absolute low of the vicious cycle when the blinders on those pet projects come off because the future numbers just won’t add up anymore. Then, large swaths of tech employees are jettisoned and, with the exception of very rare moments of candidness (if not honesty), the CEO points the finger at the “new nimble” of AI.  And then the vicious cycle repeats when CEOs like Click-Up’s Zeb Evans (also mentioned by Julie Bort) paper over the holes with talk of offering million-dollar salaries for people to babysit AI and a loose promise that they get to keep their job even if they use AI to make their own job irrelevant. Huh. Maybe it is a psychosis. So why wasn’t everyone calling out AI psychosis sooner?  Well, maybe because so many people had contracted it that it looked like “the new normal.” How to Know If You’re Suffering From AI Psychosis I asked four of my tech CEO colleagues if they were suffering from AI psychosis. Here are their answers:  “No.” “No.” “WTF?” And “No.” But here are some of the warning signs: If the CEO or their HR team are hiring for labels and prestige instead of talent, they might have AI psychosis. If the CEO is operating under the impression that they need to fire everyone because everyone else is firing everyone, or, if they’re prescient, the CEO assumes everyone will eventually be forced to fire everyone, they might have AI psychosis. If they’re throwing every last bit of cash at building AI infrastructure without questioning that if everyone else is also doing this, that house of cards cannot stand, they might have AI psychosis.  Also, look for a rash just below the nape of the neck. I’m kidding. It’d just be funny to see a bunch of CEOs having someone check their necks. How to Beat AI Psychosis The great thing about a vicious cycle is it’s not only curable, it’s preventable. Don’t start the cycle. We’ve learned that bolting AI on SaaS is a common way to start the cycle. Instead, take the preventative step and understand that the entire business process that aggregates and barfs back data to the user is over.  Don’t feed the cycle. There are a lot of psychological factors with any technical leap, and until you see symptoms—like NFTs—those factors can be hard to detect. There’s only so much money to throw at AI, so be careful where you throw yours. Don’t get caught in the cycle. And then finally, take the time to think about unintended consequences of everyone trying to be the furthest ahead in a technical evolution that is still maturing. Instead, be patient, wait for the incumbents to make their mistakes, and when the asteroid takes out the dinosaurs, start drilling for the oil they leave behind.  These are still exciting times in tech. In fact, they can be even more exciting as hype starts to give way to logic and panic starts to cede to capitulation. Don’t look at this behavior as AI psychosis. It’s insulting to people suffering from real psychoses. Instead, look at it as the same type of opportunity that comes with every tectonic technical shift. Because these massive changes aren’t “happening” to the tech industry. The tech industry is doing it to itself. For you younger kids, I’m like David Duchovny in Zoolander. Either way, join my email list and get a heads-up when I mumble something that will be relevant months from now.