Every week, there's a new headline about AI taking jobs. "AI is hitting the labor market like a tsunami," said IMF Managing Director Kristalina Georgieva. Mercer's latest survey shows employee anxiety about AI-driven job loss jumped from 28% in 2024 to 40% in 2026. And if you spend any time on social media, you'd think nobody will have a job by next year.

I've been thinking about this a lot — both as someone who builds AI tools and as someone who works with teams navigating this transition. The reality is more complicated, and more interesting, than either the doom-and-gloom or the techno-optimist narratives suggest.

Future of Work The impact of AI on work is real, but the narrative is more nuanced than headlines suggest

What the Data Actually Shows

Let's start with what we know, not what we fear.

The Layoff Picture

Yes, some companies are laying people off and citing AI as the reason. But Deutsche Bank analysts raised an important flag earlier this month: "AI redundancy washing will be a significant feature of 2026." That means some companies are using AI as a convenient justification for layoffs that were going to happen anyway — due to economic conditions, over-hiring during the pandemic boom, or strategic restructuring.

That doesn't mean AI isn't changing work. It is. But the relationship between "company adopts AI" and "company lays off workers" isn't as direct as it seems.

The Jobs Being Affected

The roles most impacted aren't the ones people usually worry about:

Most Affected Less Affected Than Expected
Data entry and processing Software engineering
Basic customer support (L1) Creative professionals
Translation (standard documents) Healthcare workers
Simple report generation Skilled trades
Bookkeeping and basic accounting Sales and relationship roles
Routine legal review Strategic consultants

The pattern is clear: AI is best at replacing routine, predictable tasks — not entire jobs. Most jobs are a mix of routine and non-routine work, so what's happening is more about task displacement than job displacement.

The Numbers in Context

The World Economic Forum's latest report says AI will create 170 million new roles globally while displacing 92 million. That's a net positive of 78 million jobs. But that aggregate number hides a messy transition: the people losing jobs aren't necessarily the same people getting the new ones.

And here's the stat nobody talks about enough: companies that adopt AI are, on average, growing faster and hiring more total people than those that don't. AI isn't a zero-sum game for most organizations. It's a productivity multiplier that often leads to expansion, not contraction.

What I'm Seeing on the Ground

Let me share what I'm actually observing in the companies I work with.

The "Do More with the Same" Pattern

Most companies aren't using AI to fire people. They're using it to get more output from existing teams. A marketing team that used to produce 10 blog posts a month now produces 30 — same headcount, more output. A customer support team handles 40% more tickets without hiring additional agents.

The nuance is this: AI is preventing hiring, not causing firing. If your company would have hired 5 more people without AI, and now they don't need to — that's a real economic effect, but it doesn't show up in layoff statistics.

The Skill Shift

I've watched the skills that matter change in real-time. Two years ago, knowing how to write SQL was a valuable skill. It still is, but now it's equally valuable to know how to tell an AI to write the right SQL — and to verify that what it wrote is correct.

The jobs aren't disappearing. The job descriptions are changing. And the pace of that change is uncomfortable for a lot of people.

The Anxiety Gap

Something I keep noticing: the people most anxious about AI are often not the ones most at risk. Senior professionals who've built deep domain expertise are actually well-positioned — they can leverage AI as a force multiplier. It's the people in the middle — doing competent but undifferentiated work — who have the most legitimate reason to think about adaptation.

But anxiety doesn't follow risk. It follows attention. And AI gets a lot of attention.

What "AI Redundancy Washing" Actually Looks Like

This is worth spending a moment on because it's a real phenomenon. Here's how it works:

  1. Company was planning restructuring due to slowing growth
  2. Leadership decides to cut 15% of workforce
  3. PR team frames it as "investing in AI to improve efficiency"
  4. Media reports "Company cuts 500 jobs due to AI"
  5. Employees at other companies read the headline and panic
  6. Survey says "40% fear losing their job to AI"

I'm not saying every AI-related layoff is fake. Some are genuine — entire categories of work really are being automated. But the narrative has gotten ahead of the reality, and that's causing disproportionate anxiety.

The Industries to Watch

Already Transforming

Financial Services. Routine analysis, fraud detection, compliance checking, and customer service are all heavily AI-assisted now. Junior analyst roles have shrunk, but demand for people who can work with AI tools has grown.

Media and Content. AI generates drafts, handles localization, and produces variations at scale. The role of a writer has shifted from "produce 5 articles a week" to "produce, curate, and quality-control 25 articles a week with AI assistance." Output up, headcount flat.

Customer Support. L1 support is increasingly automated. L2 and L3 support roles are more important than ever because the easy stuff never reaches them — they only deal with complex, unusual cases.

Slower Than Expected

Software Engineering. Despite all the AI coding tools, demand for software engineers remains strong. AI helps engineers write code faster, but the hard parts of engineering — understanding requirements, making architectural decisions, debugging complex systems — still need humans. If anything, AI is making more software projects feasible, which creates more demand for engineers.

Healthcare. AI assists with diagnosis, imaging, and administrative tasks, but the regulatory environment, liability concerns, and the fundamentally human nature of care mean the transition is deliberate and slow.

Education. Teachers are using AI as a tool, but the in-person, relationship-based nature of education is proving resistant to automation. The roles are changing (more facilitation, less lecturing), but they're not disappearing.

What Workers Should Actually Do

I've been asked this question dozens of times in the last six months, and here's the honest answer I give:

1. Learn to Work with AI, Not Against It

This isn't optional anymore. Whatever your role, there's likely an AI tool that can make you more productive. Learn to use it well. The people who thrive will be the ones who treat AI as a capable junior colleague — useful, but needing guidance and verification.

2. Double Down on What AI Can't Do

Judgment. Relationships. Creative problem-solving. Empathy. Leadership. Strategic thinking. These are the skills that become more valuable as routine tasks get automated. If you can combine deep domain expertise with AI fluency, you're in an excellent position.

3. Build a Learning Habit

The tools are changing every few months. You don't need to master every new model, but you need to stay current on what's possible. Spend an hour a week experimenting with new AI tools relevant to your work.

4. Don't Panic, But Don't Be Complacent

The worst strategy is to ignore AI and hope it goes away. The second worst strategy is to panic and make drastic career changes based on speculation. The right strategy is steady, intentional adaptation.

What Employers Should Do

Invest in Upskilling

Mercer's data is stark: 97% of investors said funding decisions would be negatively impacted by firms that fail to systematically upskill workers on AI. This isn't just a nice-to-have. Investors are watching whether you're preparing your workforce or just hoping for the best.

Be Honest About the Timeline

Don't tell your team "AI won't affect your job" if it might. And don't announce "AI is replacing your department" before you've actually validated that. Give people realistic information so they can prepare.

Redesign Roles, Don't Just Eliminate Them

When AI automates part of a job, the answer isn't always to cut the headcount. Sometimes the answer is to redesign the role to focus on higher-value work that AI enabled. The bookkeeper who used to spend 80% of their time on data entry can now spend 80% of their time on financial analysis — and they're more valuable than ever.

Create Safe Spaces to Experiment

People need permission to fail when learning new tools. If your culture punishes mistakes, nobody will experiment with AI, and you'll fall behind while your competitors adapt.

My Honest Take

I think the truth is somewhere between "AI will take all our jobs" and "AI will just create new jobs for everyone." The transition is going to be messy, uneven, and different across industries and regions. Some people will genuinely struggle. Others will thrive.

What I'm most concerned about isn't the technology — it's the speed of adaptation. AI capabilities are advancing faster than most organizations and individuals are adapting. Closing that gap is the central challenge of the next few years.

And the companies that handle this transition well — that invest in their people, are honest about the changes, and build AI into their operations thoughtfully — will be the ones that come out ahead.

Resources

Navigating AI's impact on your team or organization? CODERCOPS can help — we work with companies on responsible AI integration that empowers teams rather than replacing them.

Comments