5 Signs Your Job Is at Risk from AI (And What to Do About It)

Most people won't see AI career disruption coming. Not because the signs aren't there — but because the signs don't look like what they expect. They're waiting for a robot to show up at their desk. What actually happens is quieter, slower, and far more devastating.

Here are five warning signs that your role is on AI's hit list — and what to do about each one before it's too late.

1. Your Work Can Be Described in a Prompt

Take your last five work tasks. Now try describing each one as a ChatGPT prompt. If you can — if the instructions you'd give a competent AI are essentially the same instructions you receive from your manager — you have a prompt problem.

"Write a blog post about X for Y audience." "Analyze this spreadsheet and create a summary." "Draft a response to this customer complaint." "Create a social media calendar for next month."

Every one of those is a prompt. And every one is something AI already does competently.

What to do: Identify which parts of your role are "prompt-describable" and which require judgment, context, or relationship skills that can't be captured in instructions. Deliberately shift your time toward the latter. If 80% of your work is prompt-describable, that's your emergency signal.

2. Your Company Is "Exploring AI Tools" for Your Department

When management starts talking about "AI-powered efficiency" in your area, they're not making conversation. They're running the math. The calculation is simple: if one person with AI tools can do what three people did before, the other two become "restructuring opportunities."

The pilot program isn't a compliment. It's a tryout — and you might not be auditioning for the role you think you are.

What to do: Don't resist the AI tools. Become the person who masters them. The person who runs the pilot, who understands the tools' limitations, who bridges between AI capability and business needs — that person survives. The person who says "my work is too nuanced for AI" gets replaced by someone who disagrees.

3. You Haven't Learned a Fundamentally New Skill in 3+ Years

This one is insidious because it feels like stability. You're good at your job. You've built expertise over years. Why change?

Because AI isn't just automating current tasks — it's compressing the value of accumulated expertise in routine domains. The accountant who spent 15 years mastering tax code intricacies is now competing with AI that knows all of tax law instantly. The expertise advantage is shrinking.

The question isn't "how good am I at what I do?" It's "how good am I at learning to do new things?" The rate of change is the new metric.

What to do: Pick one skill outside your core competency and invest 5 hours per week in it for the next 90 days. Ideally, pick something that combines human judgment with technical capability — data literacy for non-technical roles, business strategy for technical roles, client relationship skills for back-office roles.

4. Your Output Is Digital and Your Input Is Digital

If your entire workflow lives on a screen — information in, processed information out — you're operating in AI's home territory. Digital-in, digital-out is exactly what AI was built for.

Compare this to a nurse, a plumber, or a trial lawyer. Their work involves physical presence, human interaction, real-world complexity, and unpredictable situations. AI can assist these roles, but it can't replace them because the core work requires being there, in the room, with a body.

The Distance Test from the book captures this precisely: roles with high "physical distance" requirements — where someone has to literally travel to receive the service — are inherently safer than roles where everything happens through a screen.

What to do: Find ways to add physical, relational, or in-person elements to your role. Lead workshops instead of writing reports. Visit clients instead of emailing them. Build the parts of your job that require presence — the one thing AI fundamentally cannot provide.

5. There's a Clear "Right Answer" to Most of Your Decisions

If your job involves making decisions where there's objectively a correct choice — optimizing ad spend, categorizing data, routing support tickets, approving standard applications — AI will eventually make those decisions faster and more accurately than you.

The safe zone is ambiguity. Roles where the right answer depends on context, relationships, values, and trade-offs that can't be quantified. Where reasonable people might disagree and the "answer" requires wisdom, not computation.

A therapist deciding how to approach a patient's grief. A creative director choosing a brand direction. An architect balancing aesthetics, function, budget, and client personality. These decisions can't be optimized because there's no single objective function.

What to do: Move toward the ambiguous end of your field. If you're in finance, move from analysis to advisory. If you're in marketing, move from execution to strategy. If you're in tech, move from implementation to architecture. The messier and more human the problem, the safer you are.

The Bottom Line

None of these signs mean your job disappears tomorrow. AI disruption is gradual — then sudden. The roles shrink first. Budgets tighten. Headcounts drop by one, then two. And then one morning the entire department is "restructured around AI-powered workflows."

The people who see the signs early have 12-24 months to reposition. The people who don't see them have 12-24 months of false security followed by a very unpleasant email.

The window is open. It won't be forever.

Not sure where you stand? Take the free AI Career Risk Audit — it takes 5 minutes and scores your career across 4 dimensions.