What will this mean for my job?

Most writing about AI and work is trying to sell you something — either the dream or the panic. If you’ve watched a demo, or seen a colleague do in four minutes what used to take you an afternoon, and felt your stomach drop, this is for you. That feeling is not naivety and it’s not you being a Luddite. It’s a sensible response to a real change, and it’s worth thinking through properly rather than waving away.

Three parts: what’s actually happening, what it does and doesn’t mean for your job, and what to do about it.

First, the fear is reasonable

When people say “AI will create more jobs than it destroys,” they are usually right about the total and quiet about the transition. Both things can be true: the economy ends up with more work for people to do, and a specific person in a specific role has a genuinely hard year. Averages don’t pay anyone’s mortgage.

It’s also reasonable to resent how it arrives. Nobody asked you whether you wanted your job redesigned. A tool shows up, a target gets raised, and suddenly the thing you were good at — the careful, experienced version of the work — is being compared to a machine that does a rougher version instantly. That’s not just an economic worry. It’s about identity, status, and the quiet pride of being the person who knows how things are done. Pretending that doesn’t matter is how you end up with a workforce that nods along in the meeting and digs in everywhere else.

So if you’re anxious: you’re reading the situation correctly, not failing to be a team player. Good. Clear eyes are the starting point, not the problem.

What’s actually happening to work

AI is not, for the most part, taking whole jobs. It’s taking tasks. Almost every job is a bundle of tasks, and AI is very good at some of them and useless at others.

It’s strong at the things that are language-shaped and repeatable: drafting, summarising, triaging, reformatting, pulling an answer out of a pile of documents, turning messy notes into a clean first version. It’s weak at the things that need judgement, relationships, accountability and context it can’t see: deciding what actually matters, reading a room, owning a decision, knowing which rule to break, being the person a client trusts.

Most roles are a mix. A salesperson spends real time on admin, research and follow-up emails — all very automatable — and real time on building trust and closing, which isn’t. An operations manager spends hours assembling reports and chasing updates, and hours making judgement calls that no model should make for them. When AI lands on a job, it tends to hollow out the repetitive middle and leave the human ends: the judgement at the top and the relationships at the bottom.

That reshaping is uncomfortable, because the repetitive middle is often where people felt safe and competent. But it points clearly at where the durable work is.

So which jobs are genuinely exposed?

The pattern is clear enough: roles that are mostly the automatable tasks, with little judgement or relationship attached, are the most exposed. Pure data entry, first-line routing, routine copy, basic first-draft work done to a template. If a role is 80% “take this input, produce that predictable output,” that role is going to change a lot, and some versions of it will go.

Roles that are a genuine blend — most of them — don’t disappear. They get rebuilt around the human parts, with the repetitive middle handed to AI. The work doesn’t vanish; the shape of a good day changes. Fewer hours assembling the report, more hours deciding what the report means and what to do about it.

That’s the difference between “my job is doomed” and “my job is changing.” For most people it’s the second one — and the second one you can do something about.

The part most people miss

Underneath the headlines it’s simpler than it looks, and oddly reassuring:

In most teams, AI won’t replace you. A person who’s good at their job and fluent with AI will replace someone who’s good at their job and isn’t.

It sounds harsh, but it’s good news. The deciding factor isn’t your age, your title, or how “techie” you are. It’s whether you make the shift. And the shift is learnable. It is not reserved for people who code, or people under thirty, or people who love gadgets. It’s a skill, and skills can be acquired on purpose.

The people who struggle most aren’t the ones who find AI hard. They’re the ones who decide, quietly, not to engage with it — who wait for it to blow over, treat it as a fad, or assume their experience exempts them. Experience is an enormous asset if you point it at the new tools. It’s a liability only if you use it as a reason not to.

What to actually do about it

Worry without action just corrodes. So — what genuinely helps, roughly in order.

1. Move from spectator to user — this month. There’s a huge difference between having seen AI and having used it on your own work. Pick one real, annoying task you did this week and do it again with AI in the loop. Not a toy prompt — a real task. The fear shrinks fast once the tool is a thing you operate rather than a thing you watch. Familiarity is the antidote to dread, and it’s mostly a matter of reps.

2. Audit your own job. Write down what you actually spend your week on, task by task. Mark each one: automatable middle or human end. This does two things at once. It shows you exactly what to hand over first (the boring bits you won’t miss), and it shows you where your real value lives — the judgement and relationship work that is now more valuable, not less, because there’s more of it in your week once the admin shrinks.

3. Become the person who connects AI to the real work. Almost anyone can type a question into a chatbot. Far fewer people can connect AI to where the work actually lives — the inbox, the files, the meeting notes — and trust it with whole tasks, safely. That’s the skill that’s scarce, and it’s the one that makes you the person a team keeps. It’s also very learnable; it’s the spine of what we teach.

4. Aim to be your team’s “AI person,” not a passenger. Every team is about to need someone who gets it — who can show others, set the sensible guardrails, and turn “we should use AI” into things that actually work. Volunteering for that role is one of the safest moves you can make, because it converts the threat into your remit. You stop being someone the change happens to and become someone who helps it happen well.

5. Lean into what AI can’t do. As the repetitive work gets cheaper, the premium shifts to judgement, taste, trust, communication and accountability. Invest there deliberately. The future of most knowledge work isn’t “human vs machine” — it’s a person doing the parts that need a person, with a tireless assistant doing the rest.

For the ones who really aren’t on board

Some people reading this won’t be reassured, and that’s fair. If you’re near the end of a career, or in one of the genuinely exposed roles, or simply tired of being told to adapt again — the advice above can feel glib.

Two things. First: you don’t have to love it to be protected by it. You don’t need enthusiasm, only enough fluency to stay in control of your own work. Plenty of people who were deeply sceptical now use AI every day and would not give it back — not because they were won round by hype, but because it quietly took the worst hours out of their week. Second: the single biggest risk isn’t moving too slowly. It’s not starting at all — letting it become the thing you avoid until it’s decided about you. Starting badly beats not starting. You can be unconvinced and still get going.

The bottom line

AI is going to change a lot of jobs, and some will go — anyone who tells you otherwise hasn’t been paying attention. But for most people, in most roles, this is a transition to be navigated, not a sentence to be served — and the people who navigate it well aren’t the most technical or the most optimistic. They’re the ones who started, who pointed their experience at the new tools, and who made themselves the person their team relies on to use this stuff properly.

That person isn’t born that way. They decide to become it. The whole point of our course is to make that decision an easy one to follow through on — to take you from worried spectator to the genuinely AI-powered person in your office, on your own real work, over eight weeks. If this article put words to something you’ve been feeling, that’s the next step worth taking.