The White-Collar Bloodbath That Never Came

 

AI won’t fire everyone—but it will expose who was replaceable all along. If your job is routine, the countdown has already started.

I have heard the warnings. They come at you like sirens at midnight. The head of the IMF says AI is a tsunami. A Wall Street titan says he’ll need fewer humans. A Silicon Valley boss says half of entry-level white-collar jobs could vanish. The chorus is loud, clean, and terrifying. It sounds like the end of the office as we know it. But I don’t buy the funeral march. I’ve seen this movie before, and the ending isn’t extinction. It’s mutation.

When I walk through the modern office in my head, I don’t see a robot sitting in my chair. I see a cyborg. Flesh and code welded together. Think less killer machine, more upgraded human. The kind that runs faster, sees sharper, and still makes the call when the stakes get ugly. AI isn’t here to wipe out white-collar work. It’s here to rip it apart, task by task, and stitch it back together into something meaner and more valuable.

Start with the numbers, because fear hates numbers when they don’t cooperate. Since late 2022, America has added roughly 3 million white-collar jobs. Management, professional roles, sales, office work—all up. Blue-collar employment, flat. That alone should slow the panic. Jobs everyone said would be first against the wall are growing. Software developers are up about 7 percent. Radiologists, up 10 percent. Paralegals, up a wild 21 percent. If AI were the executioner, these roles would be bleeding out on the floor by now. They’re not. They’re lifting weights.

Pay tells the same story. Real wages in professional and business services are up around 5 percent since late 2022. Office and administrative workers are earning about 9 percent more. Control for education, age, race, gender, the whole social autopsy, and white-collar workers still earn roughly a third more than blue-collar ones. That premium is almost triple what it was in the early 1980s. AI hasn’t stolen the crown. If anything, it’s polished it.

History backs me up, and history has a long memory. In the early 1980s, when computers started chewing through mental tasks, a Nobel Prize–winning economist warned that the bond between man and machine was being radically transformed. He wasn’t wrong. He was just early. The transformation didn’t kill white-collar work. It fed it. Since that era, employment in management, professional, sales, and office roles has more than doubled. Inflation-adjusted pay climbed by about a third. The computer didn’t erase the office. It made it bigger.

Here’s the trick everyone forgets. Technology almost never kills a job in one clean shot. It kills tasks. The boring ones. The repeatable ones. The ones that can be written as rules and fed to a machine. Typists vanished because their entire job was repetition. Most white-collar roles aren’t like that. They’re messy bundles. Analysis mixed with judgment. Coordination tangled with accountability. When computers arrived, they took the drudge work and left humans with the decisions that mattered. Productivity went up. So did wages. Air-traffic controllers didn’t disappear when software got better. They got more powerful, because someone still had to own the decision when lives were on the line.

AI follows the same pattern, just faster and sharper. Its intelligence is jagged. It’s brilliant at 95 percent of a task and shaky at the 5 percent that can blow everything up. Edge cases. Context. Discretion. That last mile where things go wrong and excuses don’t matter. Data from AI companies themselves show that only a sliver of occupations use AI across most of their tasks, and almost none can be fully automated. AI is cheapening specific cognitive moves—drafting text, scanning data, writing standard code—not entire professions.

Look at the labor market since late 2022 and you see the split clearly. White-collar employment overall is up about 4 percent. Real wages, up around three. Jobs that blend technical skill with oversight and coordination are on fire. Project managers. Information-security experts. Roles where you have to understand the system and also steer people through it. Employment in some of these has jumped roughly 30 percent. Care-oriented roles and jobs that demand judgment and human coordination are growing too. The only ones shrinking are the quiet back rooms. Insurance claims clerks are down about 13 percent. Secretaries and admin assistants, down around 20. Routine work gets eaten first. Always has.

And while everyone argues about which jobs will die, new ones are crawling out of the lab. Data annotators. Forward-deployed engineers. Chief AI officers. Titles that sound made up because they’re still settling into shape. Categories with names like “other mathematical-science occupations” have ballooned by about forty percent since late 2022, with real wages jumping roughly twenty percent. “Other computer occupations” are expanding fast. Business operations specialists, the people who redesign processes and keep the machine humming, have seen employment jump by nearly 60 percent. When productivity explodes, the economy doesn’t shrink. It sprawls.

I’m not pretending this is painless. There’s blood on the floor, and more coming. Entry-level roles are exposed because they often live on routine. New AI models can already work autonomously for hours, chaining together analysis, coding, and tool use with minimal supervision. Benchmarks show that this endurance doubles roughly every seven months. That’s not a rumor. That’s a clock ticking. Clerical and administrative work has been shrinking for decades anyway, falling from about 18 percent of the workforce in the 1980s to around 10 percent today. AI will keep pushing that number down. Workers in these roles often have fewer transferable skills and less room to climb. When the floor drops out, the fall hurts.

Still, pain isn’t apocalypse. This isn’t mass white-collar extinction. It’s selective pressure. Jobs that rely on judgment, coordination, and accountability will keep a premium because someone has to answer when things go sideways. AI doesn’t go to court. It doesn’t take the blame. It doesn’t sign off. I do. And as long as that’s true, I’m not obsolete.

I hear the panic in the break room. “So what happens to us?” someone asks. I shrug. “Same thing that always happens,” I say. “We adapt or we disappear.” It sounds harsh, but it’s honest. AI is a force multiplier, not a grim reaper. It will redraw the office again, the way computers and the internet did before. The weak tasks will die. The strong roles will evolve. New ones will be born without names. The office won’t be wiped out. It’ll be rebuilt, faster and stranger, with humans still in the loop, still on the hook, still valuable.

The future isn’t a robot takeover. It’s a negotiation. And history says the humans who show up ready to change usually walk away with the better deal.

 

 

Separate from today’s article, I recently published  “Forget Day Trading: Build Wealth the Benjamin Graham Way and Retire Rich”  for readers interested in a deeper, standalone idea. You may also read it from here on Google Play: Forget Day Trading.