Career Zone Analysis

Working out which parts of your job are exposed, which are safe, and what to do about it — without either panic or denial.

Analyse tasks, not jobs

"Will AI replace [job]?" is the wrong question and produces useless answers. Jobs are bundles of tasks, and the tasks are affected very differently.

Break your role into tasks first. Be specific — "communication" is not a task; "writing weekly status updates for the client" is.

Task	Hours/week	What it actually requires

The three zones

Score each task:

Zone 1 — Exposed. Mostly producing text, summarising, formatting, retrieving information, routine analysis. AI does a competent version today.

Zone 2 — Augmented. Requires judgement, but the production work around it is exposed. AI makes you faster; it does not do the task.

Zone 3 — Protected. Requires physical presence, accountability, relationships, or decisions someone must own. Not "AI can't do this" — "someone must be answerable for this".

Task	Zone	Why

Reading your result

Mostly Zone 1. The most exposed position, and the most actionable. The move is not to defend those tasks — it is to become the person who automates them and moves up to Zone 2. That is a real, common, successful transition.

Mostly Zone 2. The most common and the most comfortable. Your leverage rises substantially. The risk is complacency: your Zone 2 depends on judgement that needs to stay sharp, and if you stop doing the underlying work entirely, the judgement decays.

Mostly Zone 3. Least exposed, least augmented. The risk is the opposite — being slow while others get faster on the parts around your work.

What actually protects a career

Not "skills AI cannot do". That list keeps shrinking and anyone selling it is guessing.

What holds:

Accountability. Someone must be answerable. AI cannot be sued, fired, or trusted. Roles where a person owns the outcome do not disappear.

Relationships and context. Knowing which colleague to ask, why the last project failed, what the client actually meant — this is not written down anywhere for a model to learn.

Judgement under ambiguity. Deciding what to do when the situation is not in the training data.

Taste. Knowing which of five competent options is right for this audience. Hard to articulate, very hard to automate.

Being the person who builds the leverage. In every organisation there are a few people who figure out how to make the tools work. That position is currently undersupplied and well rewarded.

The honest self-assessment

Hours/week in Zone 1
What I would do with those hours
The judgement my Zone 2 depends on
How I keep that sharp
Who I am to my colleagues, beyond output
What I know that is not written down

Row four matters more than people expect. If AI does all the drafting and you only ever review, your ability to judge the drafts erodes quietly.

The three moves

Automate your own Zone 1. Do not wait to be asked. The person who automates their own routine work and reports the saving is in a strong position; the person whose work is automated by someone else is not.

Make the leverage visible. Quantify it. "I cut our reporting cycle from 3 days to 4 hours" is a career asset; "I use AI a lot" is not.

Deepen one thing. Broad AI familiarity is becoming common and therefore worth less. Combining it with genuine depth in your domain is not.

Conversations to have

With your manager, once you have a real result:

I automated [TASK]. It went from [BEFORE] to [AFTER].
I would like to use some of that time on [HIGHER-VALUE WORK].
Can we agree what that should be?

Lead with the result, then the ask. This conversation goes very differently from "I've been learning AI".

With yourself, quarterly:

• What did I do this quarter that only I could have done?
• What did I do that a competent person with good tools could have done?
• Is the first number growing?

The thing not to do

Do not build your position on being the only person who knows how to use the tools. That advantage has a short half-life — the tools get easier and everyone catches up.

Build it on judgement about what to build and why, which does not.

Review

Quarter	Zone 1 hours	What I automated	What I moved up to

If the "moved up to" column is empty, the automation saved time you then gave back to more Zone 1 work. That is the trap, and it is easy to fall into.