Remember that viral Substack post that had everyone panicking about AI turning us all into unemployed couch potatoes by 2028? Yeah, well, Citadel Securities just stepped into the chat with a massive “hold my beer” moment.
So here’s what happened: Some research outfit called Citrini published this dystopian fanfiction called “The 2028 Global Intelligence Crisis” over the weekend. It painted a picture so grim it would make a horror movie director jealous – 10.2% unemployment, S&P 500 down 38%, basically AI eating everyone’s lunch and leaving us with the crumbs.
Wall Street, being Wall Street, immediately freaked out and started selling everything that wasn’t nailed down. Because apparently we’re all just one scary blog post away from a full-blown panic attack.
Enter Frank Flight, Citadel’s macro strategist, who basically said “Not so fast, drama queens” and dropped some actual data to cool everyone’s jets.
The Reality Check We All Needed
Flight started his rebuttal with some real numbers that sound way less apocalyptic: “The year is 2026. The unemployment rate just printed 4.28%, AI capex is 2% of GDP, and – plot twist – job postings for software engineers are actually UP 11% year-over-year.” Oops, there goes that narrative.
Turns out, when you look at actual Federal Reserve data instead of science fiction, AI isn’t exactly steamrolling the job market. In fact, the data on daily AI use for work is “unexpectedly stable” – which is finance-speak for “nothing to see here, folks.”
Why the Robot Takeover Is Running Behind Schedule
Here’s where it gets interesting. Everyone’s been assuming AI adoption would be like flipping a switch – one day you’re doing your job, the next day a robot is sitting at your desk wearing your favorite coffee mug. But reality is messier than that.
Citadel points out that AI adoption is actually expensive and slow. Shocking, I know. Turns out integrating AI into real businesses involves things like “integration costs,” “diminishing returns,” and “regulatory costs” – basically all the boring stuff that makes tech rollouts take forever.
Plus, there’s the small matter of AI needing massive amounts of computing power, which – surprise! – is in short supply. We’ve got chip shortages, data centers that take forever to build, and energy constraints that would make a Tesla owner nervous. Even if AI gets smart enough to replace us, good luck finding the infrastructure to actually do it at scale.
Economics 101: Why Productivity Booms Are Actually Good
Here’s the kicker that the doomsday crowd seems to have missed: productivity shocks are usually good for the economy. I know, wild concept. When companies can produce more stuff for less money, prices fall, margins expand, and people have more purchasing power. It’s almost like every major tech advancement in history – from steam engines to the internet – has made life better, not worse.
The bears keep insisting “this time is different” because AI might directly replace workers. But Flight basically said that’s like claiming companies will produce more while demand collapses – which violates basic accounting principles. In other words, it doesn’t add up.
The Bottom Line
Look, AI is definitely going to change things. But maybe, just maybe, we don’t need to panic every time someone publishes a scary story about robot overlords. The data suggests we’re nowhere near the AI apocalypse, and the infrastructure constraints alone mean any major disruption is going to happen gradually, not overnight.
So the next time you see a viral post predicting the end of work as we know it, maybe take a deep breath and remember: Wall Street has been wrong about the future before. Like, a lot.
Now if you’ll excuse me, I need to go update my LinkedIn profile to “AI-Resistant Human” just in case.