AI in the Workplace: IBM’s Double-Edged Bet

So here’s the setup: IBM swapped out a few hundred HR folks for AI systems. No scandal. No apology tour. Just a straight-faced statement from the CEO saying, “Yeah, we did that. But hey—we’re hiring elsewhere, so… balance?”

Welcome to the new corporate chessboard, where some jobs get moved up, others get quietly taken off the board, and the name of the game is "strategic efficiency." Depending on who you ask, IBM is either leading the charge toward a smarter future—or slowly unraveling the safety net workers once counted on.

What’s clear is this: AI didn’t just sneak into the workplace. It kicked the door in, helped itself to your swivel chair, and now it’s answering emails faster than you can spell “redundancy.” But before you reach for the panic button, consider this—IBM claims that overall hiring has actually gone up. Different jobs, sure. But more jobs? That’s their line.

Still, let’s not get too cozy. Because behind the smooth messaging and polished stats is a question that won’t go away: what happens to the people left behind?

This piece digs into that question—looking past the headlines, beyond the boardroom bravado, and into what IBM’s AI play really means for the rest of us who still bleed coffee and live by the Outlook calendar.

IBM’s HR Revolution: When Bots Start Handling People

In early 2025, IBM made a move that felt straight out of a Black Mirror subplot: they handed routine HR tasks to artificial intelligence and told hundreds of employees their services were no longer required.

The tasks replaced? Not flashy. Processing vacation requests. Answering the same ten questions about benefits. Managing pay statements. The kind of work that lives in the background—until it's suddenly gone and so are the people doing it.

CEO Arvind Krishna didn’t spin it. He told The Wall Street Journal that several hundred roles had already been replaced by AI. But he was also quick to frame it as a win. The money saved? Reinvested in areas that “require critical thinking and human interaction”—software engineering, sales, marketing.

And that’s the pivot: IBM isn’t anti-people. They’re anti-redundancy. If a task can be done better, faster, and cheaper by a machine, they’ll make the swap—and they’ll tell you about it with a smile.

Let’s be honest: it’s not a mass firing. It’s a selective edit. Out with the manual tasks, in with the roles that can’t (yet) be automated. According to internal stats, IBM’s AI-driven HR systems handled 94% of 11.5 million HR queries. That's not a minor upgrade. That’s a pitch deck headline.

But while the company talks about productivity gains and job reallocation, the folks who lost their jobs tell a different story. Minimal notice. Slim severance. Few options. The retraining window was barely cracked open before the door shut behind them.

And that’s where the story gets sticky. Because AI might not be evil, but it sure as (insert expletive of your choice) isn’t sentimental.

Krishna isn’t shy about what’s next either: he says thousands more roles could go in the coming years. That message has echoed through IBM’s halls, stirring up anxiety in departments far beyond HR. Sure, there’s an “AI Readiness” program to help employees reskill. But it’s voluntary. And let’s be real, whoever’s left is probably juggling two jobs just to keep the wheels turning. Now ask yourself: when exactly are they supposed to prep for a third?

Bottom line? IBM’s approach is calculated, deliberate, and unapologetic. Automate where you can. Redirect talent where it matters. And if some people get left behind… well, that’s progress, isn’t it?

Everyone's Playing IBM’s Game Now—Some Just Aren’t Admitting It

IBM might’ve made headlines, but don’t let that fool you into thinking they’re the outlier. What they did? That’s the blueprint now. In the first few months of 2025 alone, more than 50,000 tech jobs vanished. And the reason companies gave? AI.

UPS slashed 20,000 jobs—its biggest layoff since horses pulled their delivery wagons. The company leaned hard into AI-powered logistics, streamlining operations like a spreadsheet come to life. Unlike IBM, they didn’t sugarcoat it with promises of "reinvestment." No shiny talk of upskilling or talent redeployment. Just cuts.

Over in Silicon Valley, the big players lined up like dominos. Microsoft let go of 6,000 people—3% of its global workforce—under the banner of “AI-first operations.” Translation: if a bot can do your job, don’t bother logging in. Google and Meta followed suit, trimming fat in mid-level engineering roles. It’s open season on jobs that live in predictable code and pattern recognition.

Even the banks, usually the last ones to blink, are quietly pulling the same levers. No big announcements, no dramatic pink slips—just a slow bleed. Customer service, compliance, operations? All slowly being handed to machines that don’t take coffee breaks or file lawsuits.

Want a CEO who skips the PR dance entirely? Say hello to Klarna. Their chief exec flat-out bragged that AI replaced 700 customer service agents, and he didn’t bother wrapping it in feel-good jargon. “Same volume of inquiries, fraction of the cost,” he said. No reallocation, no retraining. Just subtraction.

So here’s the pattern:

• AI eats the repetitive jobs first—admin, HR, customer service.

• Some companies dress it up as progress. Others just tighten the belt.

• But no matter how you spin it, the net result so far has been fewer people and more machines.

And workers? They’re catching on. A recent survey showed that 30% of employees worry AI is coming for their job. Not because they saw a sci-fi movie—but because they’re paying attention. Entry-level knowledge work, customer support, back-office admin—these aren’t safe bets anymore.

The World Economic Forum didn’t pull punches either. In their “Future of Jobs” report, 40% of companies admitted they’re planning to reduce headcount where AI can step in. Sure, most of them say they’re investing in training too—but let’s be honest: that doesn’t help the person whose job disappeared last Thursday.

This isn’t paranoia. It’s not hype. It’s restructuring—slow, steady, and relentless.

The real question isn’t “Will AI replace jobs?” That ship has sailed. The real question is:
Who gets replaced, how fast, and what kind of safety net—if any—do they get on the way down?

When AI Goes Off the Rails

Let’s be clear: AI can be brilliant. But when it’s wrong? It’s wrong fast, loud, and sometimes- downright illegal.

Case in point: a couple of lawyers used ChatGPT to write a legal brief. The tool invented six fake court cases out of thin air. And when the judge figured it out? Sanctions. No robot scapegoat. Just human consequences. For giggles, look up Mata v. Avianca (2023).

Air Canada ran into its own AI debacle when a chatbot misinformed a grieving customer about bereavement fare refunds. He followed its instructions. The airline denied the refund. A court ruled the AI’s bad advice was still the airline’s responsibility. Lesson learned: If your AI talks like it knows the policy, you’d better make sure it does.

Australia saw a real estate listing that hyped up a house for its proximity to two great schools… that don’t exist. Blame the AI. Blame the humans who didn’t double-check. Either way, credibility got torched.

Even in science, where facts are supposed to matter most, AI snuck fake images into published research journals. At CNET, AI-generated financial articles were riddled with errors. The company had to hit pause, issue corrections, and clean up the digital mess.

In every case, the theme is the same: the AI didn’t ruin the moment. The humans not watching it closely enough did.

Now, apply that to HR. Imagine a bot miscalculates severance. Or sends the wrong termination notice. Or gives incorrect benefits info. That’s not just a glitch—that’s a lawsuit waiting to happen. And the press will eat it alive.

Efficiency is great. But when you cut corners on oversight, you’re not saving money. You’re gambling with your reputation.

What This Does to Your Employer Brand

Let’s say you’re a company that went all in on AI. You trimmed the fat. Your numbers look great on paper. Cool.

But here’s the part that doesn’t show up in quarterly reports:
Nobody trusts you anymore.

I imagine anxiety among IBM employees spiked after the HR shake-up. The company saying morale is fine, but whispers saying otherwise. You can’t tell people “you’re next” with a smile and expect high engagement scores in the next survey. Sigh.

Survivors—the ones still employed—aren’t celebrating. They’re looking over their shoulders. Productivity? Down. Innovation? Stalled. Why build the future for a company that might replace you tomorrow?

Ironically, these same companies later struggle to hire the very people needed to manage or work alongside the AI systems they installed. Trust erodes. Talent walks. The long game starts to look more like a slow leak.

Some companies are trying to do it better. Salesforce committed to retraining employees before introducing AI-driven changes. Microsoft built an AI ethics committee to vet decisions from both a business and human lens. Smart moves. Not just because they’re “nice,” but because they’re strategic. Trust is currency. And once you burn it, it’s hard to buy back.

So What’s the Playbook Now?

If your company is rolling out AI—or even thinking about it—don’t just think about the tech. Think about the people. And not just the ones coding it.

Here’s what the grown-ups are doing:

• Human oversight, always. Someone needs to check the AI’s work—especially in HR, legal, finance, and customer-facing roles.

• Communicate like you mean it. Don’t let employees find out from a press release. Spell it out. Timeline. Goals. Impact. Honesty builds trust.

• Plan before you pull the trigger. Have your reskilling programs ready before you drop pink slips. Make internal mobility real, not just a buzzword.

• Think long game. Your brand, your hiring pipeline, your internal culture—they all take hits when you treat people like widgets.

• Use AI to augment, not erase. Let it handle the grunt work, and free up humans to do what only humans can—think, create, connect.

Because here’s the deal: the companies that survive this shift won’t be the ones with the flashiest tech. They’ll be the ones who lead with intelligence—artificial and otherwise.

Final Word: The Edge Cuts Both Ways

IBM’s play was bold. Strategic. And a little cold. But that’s the thing about AI—it’s a tool. A mirror. It reflects the priorities of the people using it.

Handled well, AI can elevate human potential. Handled poorly, it becomes a wrecking ball dressed like a spreadsheet.

So yeah—AI is here. It’s not the enemy. But it is a test.

And how we answer says a lot more about us than it does about the machines.

The Comics Section

One more thing before I go...

At the last Recruiting Innovation Summit (shoutout to the sponsors for backing this newsletter), the air was thick with one thing: AI. Couldn’t escape it. Wouldn’t even try.

But what really stuck with me wasn’t the tech—it was the people.

If you were a lifer with one foot in retirement? You were chill. Probably thinking about golf.
If you were five years or less into the game? That tight smile said it all. Panic behind the nametag.
And the folks in the middle? Asking for advice. Hunting for some “safe zone” AI hadn’t already bulldozed. Getting frustrated when it started to feel like that zone didn’t exist.

So now I’m wondering—what’s your vibe?

Are you anxious? Ambivalent? Already plotting your great escape from the industry?

Hit reply and tell me. Operators are standing by (and yes, for now, they're still human).