AI Replacing Jobs Is the Wrong Fear. Here's What's Actually Happening to Your Team.

Your people are worried. They've read the headlines. They've heard the whispers. They fear that AI is replacing jobs. They need to hear this from you - not from a news feed.

Your Team Thinks AI Is Replacing Jobs. Here's What the Research Actually Shows.

Picture your finance director on a Monday morning. She's spent 20 years building her career - deep expertise, hard-won judgment, the kind of pattern recognition that only comes from doing the work. Then a colleague mentions that AI can now draft the variance commentary she spends three hours on every month. She doesn't feel empowered. She feels expendable.

Now multiply that across your org. Your senior analyst. Your project manager. Your customer service lead. The people you rely on most are the ones reading headlines about their own obsolescence.

Anthropic just released research that gives you something concrete to say to them - and a clear picture of what to do next.

The study, "Labor Market Impacts of AI" (Massenkoff & McCrory, March 2026), built a new metric called "observed exposure" that tracks the gap between what AI could theoretically do and what people are actually using it for. That gap is enormous. And the unemployment data? No statistically significant increase for AI-exposed workers since late 2022. People who have established jobs are keeping them.

AI is augmenting your team. It is not replacing them. That's not a talking point. It's what the data shows.

The Real Opportunity Most Leaders Are Missing

Here's where the research gets actionable. Computer and math roles show 94% theoretical AI exposure - meaning AI could assist with nearly every task in those jobs. But only 33% of those tasks show actual AI usage. The researchers found that compliance rules, software limitations, and the need for human verification are the main reasons that gap exists. The most exposed roles right now are Computer Programmers, Customer Service Representatives, and Data Entry Keyers.

And here's the part that should change how you think about your org chart: the workers most exposed to AI aren't entry-level. They earn 47% more on average than unexposed workers. They're significantly more likely to hold graduate degrees (17.4% vs. 4.5%). They skew female. These are your highest-value knowledge workers — and they're the ones spending hours on tasks AI handles well.

Every hour your finance director spends formatting a report is an hour she's not spending on the judgment call that prevents a bad investment. Every afternoon your project manager spends compiling status updates is an afternoon she's not spending on the relationship management that keeps a client from churning.

The goal isn't to protect your people from AI. It's to give them the tools, frameworks, and permission to hand off the repetitive work and step into the high-value work that no AI can do — strategic thinking, relationship building, creative problem-solving, and experienced judgment.

That's what augmentation actually looks like. Not a press release. A manager sitting down with their team and saying: "Here's what AI handles now. Here's what I need you to focus on instead. And here's how we'll get you there."

Theoretical capability and observed exposure by occupational category

Figure: Share of job tasks that LLMs could theoretically perform (blue area) and our own job coverage measure derived from usage data (red area).

AI Is Absorbing Entry-Level Work. You Still Need Junior Talent - Here's How to Rethink It.

There's one finding in the research that deserves a separate conversation. Among workers ages 22–25, there's a roughly 14% drop in the rate of finding new jobs in highly AI-exposed occupations. The research suggests companies are using AI to absorb the routine tasks that used to go to fresh graduates.

On the surface, that looks efficient. Underneath, it's a pipeline problem.

Those "routine tasks" weren't just busywork. Data pulls taught analysts how the business actually runs. Drafting customer responses taught service reps how to read context and tone. Compiling reports taught coordinators how information flows across departments. The grunt work was the training ground.

The answer isn't to stop hiring junior talent. It's to redesign what those roles look like.

That means new hires learn by working alongside AI - reviewing outputs, catching what it gets wrong, understanding why the human layer matters. They get exposure to judgment-heavy work earlier because the rote work that used to build up to it is shrinking. They need a framework for when to trust the tool and when to override it.

This isn't a hiring freeze. It's a change management challenge. And the companies that solve it first will build the strongest benches.

New job starts among workers age 22-25 in occupations with high observed exposure and no AI exposure, Current Population Survey. The top panel shows the percent of young workers starting new jobs in high vs. no exposure occupations. The bottom panel measures the gap between these two series in a difference-in-differences framework.

Why AI Adoption Fails Without Change Management

Every finding in this research - the 94%-to-33% capability-usage gap, the junior talent shift, the concentration of exposure among your most valuable people — points back to the same thing. This is an adoption and change management problem. Not a technology problem.

The tools exist. The capability is there. What's missing is the structure. If you've used the BORE Framework (tasks that are Boring, Overhead, Routine, or Easy to delegate), you already know the starting point: map the tasks AI should handle, identify the higher-value work your people should move into, and build the bridge between the two. That's use-case mapping, role-by-role enablement, and practical guardrails — not a tech rollout.

Companies that treat AI adoption as a software implementation will close that 94%-to-33% gap slowly, if at all. Companies that treat it as a change initiative - with clear use cases, measurable outcomes, and people at the center - will move their teams from exposed to equipped.

Try This: The 15-Minute Exposure Audit

Pick one department. List the five most time-consuming recurring tasks - weekly reports, data pulls, customer response drafting, scheduling coordination, document review. For each one, answer three questions:

1. Is someone already using AI to do part of this? (You might be surprised.)

2. If not, could they - with 30 minutes of guidance?

3. Who used to learn the business by doing this task?

Questions 1 and 2 show you where enablement earns its keep. Question 3 shows you where your onboarding needs to be redesigned. Run this audit, and you'll have a clearer picture of your AI adoption reality than most companies twice your size.

The Leaders Who Move Now Will Win the AI Augmentation Race

We're still in the early days of enterprise AI. The Anthropic researchers plan to continuously update their framework as capabilities evolve - and so should you. Right now, the data says AI is a tool for augmenting your current employees, not replacing them. But the shift is real, it's measurable, and the gap between what's possible and what's happening is where competitive advantage lives.

Your people don't need reassurance. They need a plan. The leaders who give them one — with clarity, structure, and real momentum behind it - will be the ones who turn this from a source of anxiety into a source of strength.

Your next step: Run the 15-minute exposure audit with one department this week. Then book a conversation with us to turn what you find into a 90-day plan.

Source: Massenkoff, M. & McCrory, P. (March 2026). "Labor market impacts of AI: A new measure and early evidence." Anthropic Research. https://www.anthropic.com/research/labor-market-impacts