The AI Job Apocalypse Debate is Wildly Flawed
The AI job-pocalypse conversation is infuriating.
Consider what everyone is talking about. And I do mean everyone.
The fundamental argument is focused on historical precedent of technologies. Lump of labor! Jevons paradox! Every time technology displaced workers, new jobs emerged!
Bro.
Here's the fundamental problem, and it's alarmingly simple.
Those frameworks are all based on technology.
But AI doesn't behave like a technology. It behaves like a person. It talks like a person and "reasons" like a person. And that's in 2026. Getting exponentially faster. Now consider where we will be in five years.
So I ask you:
What's the historical precedent for a digital species landing on earth that is getting closer and closer to doing what we do?
Listen - I'm bullish on humanity here. And I believe there is a solution. But for the love of Pete we really need to start focusing on the right stuff.
WHY THE OLD PLAYBOOK IS LESS APPLICABLE
There are two AI's:
There's the AI built into products. That's technology. But the other AI is an augmentation of how everyone in your company thinks. And not just inside the office but everywhere in life.
We're conflating the two.
Every time a new technology showed up and people freaked out about jobs, the same thing happened: machines took over the doing, and humans moved up into the thinking. Physical labor to knowledge work. Routine stuff to creative stuff.
That's been the escape valve every single time. And it's worked every single time, because no technology could follow us there. Machines couldn't do that cognitive lifting.
But now this is sneaking up on us.
Ken Griffin at Citadel says it's doing PhD-level work at one of the world's largest and most sophisticated financial institutions. This is something that is increasingly mirroring human cognition, and we're out here asking frameworks designed for tractors and spreadsheets to tell us what happens next.
And here's what makes this even weirder to predict: people don't particularly like AI. There's resistance. Backlash, even.
But you know who likes it?
Companies. And people who are actually using it - the ones who've made their work and lives dramatically better with it.
That tension is where the real story is.
THE REAL DIVIDE
The job apocalypse - if it comes - won't look like what people think.
It won't be AI replacing humans. And it won't even be that tired line about "you'll be replaced by someone using AI." That puts all the burden on the individual, as if tinkering with ChatGPT on your lunch break is going to transform an organization.
It doesn't work that way.
When somebody gets really good at AI, starts producing incredible work, bosses simply don’t always notice. That’s because the systems and processes around them haven't changed. Their manager doesn't know what to do with it. And with no framework for scaling impact, nothing happens.
Here's what's actually going to separate winners from losers:
The companies that figure out how to rewire their operations. The ones that teach their people how to think differently, build new processes from the ground up, and trade mere productivity gains for innovation gains.
Those companies are going to do dramatically more with their people, and hopefully hire a lot more of them, because suddenly each person can do so much more.
The companies that don't will keep running the same playbook and wonder why they're falling behind.
Everyone is arguing about whether AI will take your job. Almost nobody is talking about the fact that this is an organizational behavior problem, not an individual skill problem. Companies can’t just buy a new tool. They have to fundamentally change how they operate, from the top down.
WHY I'M OPTIMISTIC
Here's the good news: I truly believe we are going to solve this.
This isn't "learn to code" or "go back to school." The barrier to using AI is laughably low. You literally just talk to it, like texting a friend who happens to know everything.
And the organizational challenge? It's hard, but it's not mysterious. We know what it takes: Teach people how to think with AI, not just how to use it. Change the processes, not just the tools, and set expectations, not just encouragement.
The companies that do this aren't just going to survive. They're going to thrive in ways that are genuinely exciting.
But we have to build a workforce that drives that change. People ARE the company. And we’re getting there, slowly. More and more folks are loving it and living it and finding the magic to change how they work.
You know these people. You might be one of them.
The gap is forming, yes, but the door is still wide open. We just need to stop having the wrong conversation and start having the right one.
Let's do this, friends.
AI NEWS THIS WEEK
Two giant banks, same week, opposite bets
HSBC told 200,000 staff to get trained and get on board; its CEO framed AI as destroying some jobs while creating others, with retraining as the mission. Standard Chartered just went the other way — cutting nearly 8,000 roles and calling it swapping out "lower-value human capital." Same tech, two boards, the divide playing out live.
ANTHROPIC’S OPUS CLIMBS
Anthropic dropped Claude 4.8. It tops the industry benchmarks for AI coding - biggest jump? Vibe coding. Non-coders out there - this one's for you. Opus 4.8 outperforms every other publicly available AI model on the vibe coding benchmark from Vals AI, a firm that tracks how the latest models perform.
The question quietly changed from "which model" to "who's in charge of the robots"
This week the enterprise story stopped being about raw model power and became about governing agents that act on their own; audit trails, permissions, a human in the loop. ServiceNow, Salesforce, and Microsoft are all racing to be the control tower. Which is the tell: the hard part was never the tool. It's running the place around it.
