thinkingIs My Job Actually Safe?
The results came back fast. Articles from three years ago: no, don't worry, human judgment is irreplaceable. Articles from last year: yes, probably, timelines vary. A Reddit thread I shouldn't have read at 11pm. And underneath all of it, something I couldn't quite name. Not panic exactly. More like the floor had shifted by a few inches. Not fallen through. Just not where I'd left it.
I've changed jobs before. I've adapted before. I'm not someone who thinks the world owes them job security. I've been around long enough to know that industries shift, companies restructure, skills become obsolete. What felt different about this was the speed. I wasn't watching something change across a decade. I was watching it change across a product release cycle.
The Google results weren't wrong. They just weren't answering the question I was actually asking. Which wasn't "will my profession survive." It was "will I, specifically, in this specific role, still be useful to someone in three years." Those are different questions, and I couldn't find anyone asking the second one honestly.
A friend of mine, a paralegal, fifteen years, genuinely excellent at it, told me something a few months ago that I've been turning over ever since. Her firm had started using AI for document review. Not replacing anyone, they said. Just augmenting. Her first reaction was relief. The document review work was the part she liked least. Tedious. The AI was welcome to it.
Then she noticed something. The junior associates, who used to spend their first two years doing exactly that work (learning the cases, learning the clients, learning how things actually run) weren't doing it anymore. Which meant they weren't learning from doing it either. The firm hadn't replaced anyone. But it had quietly removed the thing that used to turn a new person into a capable person.
She said something I haven't been able to shake: "I learned this job by doing the boring parts. I don't know how the next lot are going to learn it."
She's fine. Her job is fine. But the point wasn't about her safety. It was about what the work actually is, once you remove the parts the AI has taken. And whether what's left is the same job with different tools, or a different job with the same title.
That reframe helped. Not "will my job exist" but "what will a job like mine actually require in five years." The people who are probably most settled about this aren't the ones who've been doing the same thing for a decade. They're the ones who've been doing it and watching how it's changing at the same time.
Which sounds fine until you follow it to where it actually leads. Adapting is real work. It falls unevenly. Someone with a supportive employer, a training budget, flexible hours, and a role that gives them room to experiment will adapt differently from someone working a job that leaves no time or headroom to think about any of this. The people who most need to get ahead of this shift are often the ones with least capacity to do so. I don't know what to do with that. But I don't think it's honest to skip past it.
And there's something else I keep circling back to. The question of who decides which parts of a job are the boring parts the AI can have, and which parts are the human judgment that stays. That decision is being made right now, mostly by people who are not in those roles. That's worth sitting with, even if sitting with it doesn't immediately produce anything useful.
My father worked in print. He watched it change slowly enough that he could see it coming and make decisions. He retired before the last of it went. I've been thinking about how different that looks from where I'm sitting, where the change isn't arriving in the building, it's arriving in a software update, quietly, between versions. Roles don't disappear overnight. They shift at the task level, until one day the job has a different shape and no one announced the change.
That doesn't make it more or less alarming than it sounds. But it means the question "is my job safe" might be the wrong scale. The task is the right scale.
Questions people ask
Is it normal to feel anxious about AI and your job even when nothing has actually changed yet?
Yes, and it's worth knowing that this specific anxiety is one of the more honest responses available. The unease tends to arrive before the evidence does, and that's not irrationality. That's pattern recognition operating on a situation that is genuinely uncertain. The people who feel nothing about this are not necessarily calmer. They may just not be paying attention yet. Neither panic nor total unconcern is the right register. Somewhere in the middle, attentive, specific, not catastrophising, is probably the most useful place to be.
What if I look at my work honestly and feel more exposed than I expected to?
The most useful thing is probably not reassurance, and I think that's actually fine to admit. That feeling of exposure, when it arrives, is information. It doesn't have to be acted on immediately. Sitting with it for a bit, rather than either dismissing it or catastrophising, is probably closer to the right response than either extreme.
Should I be talking to my employer about this, or is that the kind of thing that flags me as worried?
It depends on the culture you're in, and only you know that. But in most workplaces, the people who are quietly working out how to use AI to do their job better are not the ones who look vulnerable. They're the ones who look ahead of the curve. Asking your employer how they're thinking about AI in your role is different from asking if your role is at risk. The framing matters. "I've been thinking about how some of these tools might apply to what we do. Is that something the team is exploring?" is a very different conversation from "am I about to be replaced."
Still thinking? So are we.
This stuff is hard to make sense of — and honestly, your thoughts help us do that better. If something in this post landed, or didn’t, or took you somewhere unexpected, we’d love to know. We read everything that comes in. We can’t always write back personally, but if your thought sparks something we think others need to hear, we’ll address it in a future post.
If that happens, your thought stays anonymous — always. If you’d like your initials included, just add them and we’ll take that as permission. Either way, we’ll send you the link if you left an email.
No public comments. Just an honest conversation between us.