
How do you prove your judgement in an interview when AI can do the task?
Halfway through the interview, the manager says it, half impressed and half testing: honestly, a lot of this we could get an AI to do now. The room goes still. And how you answer the next sixty seconds may decide the whole thing.
The instinct is to defend the task, to insist the thing you do is harder than it looks. Resist it. The task may well be automatable, and arguing otherwise makes you look like someone clinging to a job description. The point is the judgement around the task, and that is what you are there to prove.
So shift the story. Do not describe what you produce. Describe a decision you made.
Pick a real moment where the obvious, automatable answer would have been wrong, and you caught it. The report that was technically correct but would have landed badly with a nervous client, so you reframed it. The optimal plan that ignored a quiet politics in the room, so you traded a little efficiency for buy-in. The data that pointed one way until you noticed what it was missing.
That is the gold.
An AI can generate the first draft. It cannot feel a room go cold, weigh a relationship against a number, or carry a sceptical team to a yes. When you tell a decision story, you are showing the one thing that does not automate: context, judgement, and the nerve to override the obvious when it is wrong.
Build the answer in three beats.
- Here is what the situation looked like
- Here is the decision I made, why the easy answer would have failed
- Here is what happened because I made it. Short, specific, and about a choice, not an output
Then go one step further, because this is what truly lands now. Show that you use the tools well. I would happily let AI draft this; my value is knowing which draft to trust, what it is missing, and when to throw it out. Now you are not the person afraid of the technology. You are the person who directs it. That is a promotion in the interviewer's mind, not a threat.
Here is the quiet reframe to walk in with. The question is not really can AI do your job. It is do you understand where your value has moved to. The people who thrive are not pretending the tools do not exist. They are the ones who can point, clearly, to the judgement the tools still cannot make.
So when they raise the machine, do not flinch. Tell the story of a decision only you could have made, and the room stops comparing you to a tool and starts seeing a professional.


