Is Your Job Safe? What AI Actually Means for the Way We Work

Not every job is disappearing, but the map of what matters is being redrawn right now.

Let's be honest. When you hear that AI is coming for jobs, your stomach probably does a small, quiet flip. It doesn't matter if you're a fresh graduate or twenty years into a career; the question creeps in: Am I next?

The good news is that the picture is far more nuanced than the headlines suggest. The bad news is that some jobs really are slipping away, and the sooner you understand which ones, the better placed you are to do something about it.

So let's cut through the noise and talk about what's actually happening.

The Jobs That Aren't Going Anywhere

Think about the last time you called a plumber. Or the moment an ambulance team showed up at your door. Or the surgeon who told you, calmly, exactly what was going to happen next. There’s a reason those interactions felt deeply human, because they were. And that’s precisely why AI can’t replicate them.

Roles built on physical presence, genuine trust, and real accountability are the most insulated from automation. Firefighters, electricians, nurses, therapists, teachers, these jobs require you to be there, in the room, with your hands and your judgment and your humanity fully engaged.

The jobs least likely to be automated are the ones that require judgment or taste.

Even lawyers are safer than you'd think. You can't walk into the courtroom without passing the bar. Regulation keeps humans in the room. The same goes for medicine, where insurers still need a licensed physician on record, and where a robot operating on your spine is still a long way from being something society is comfortable with.

That doesn't mean these workers won't use AI. A firefighter might one day wear a helmet that reads heat signatures and flags structural weak points. A doctor might rely on AI to review scans before making a call. But humans are still at the center of the decision. AI augments; it doesn't replace.

The Jobs That Are Genuinely at Risk

Here's where it gets uncomfortable. If your work mostly involves sitting at a desk, processing information, filling templates, scheduling things, transcribing things, or doing tasks that are repetitive and don't require you to physically go somewhere, you're in a more vulnerable position than you might like to admit.

AI is extraordinarily good at this kind of work. It doesn't get tired. It doesn't lose focus at 3 pm. It can do in seconds what might take a human hours. Tasks like data entry, scheduling, captioning, basic research, and even parts of recruiting are already being eaten up by automation.

And it's not always a dramatic replacement, sometimes it's quieter than that. A company that once needed ten recruiters might need three, because AI handles initial screening and fewer positions need filling anyway (since existing employees, augmented by AI, are simply doing more). The headcount just shrinks.

The Jobs in Between — The Interesting Ones

Then there's a whole category of roles that won't disappear but won't stay the same either. These are the jobs that will transform, and honestly, they're the most interesting ones to watch.

Take radiology. You still need a human to guide a nervous patient into an MRI machine, to explain what's happening, and to hold the room together. But AI is getting frighteningly good at reading scans. So, the number of radiologists needed might shrink, even as the ones who remain become more powerful because of the tools they have.

Truck driving is another one. Fully autonomous fleets aren't here yet. But when the safety case tips, when the data shows robots crash less often than humans, adoption will move fast. These jobs aren't gone, but they're on a clock.

What About Tech, IT and Manufacturing?

This is the one that really surprises people. Tech workers, the very people building AI, are not automatically immune to it. In fact, some of them are on the front line of disruption in a way that's deeply uncomfortable to admit.

Take junior developers. A few years ago, an entry-level coder fresh out of university could expect to spend their first couple of years writing boilerplate code, fixing bugs, and slowly building up experience. Today, AI can generate that boilerplate in seconds. Companies are already reporting that they need fewer junior engineers because senior developers, armed with AI tools, are simply more productive than ever. The ladder is still there, but a few of the lower steps have been cut away.

That said, software engineers who can architect systems, lead teams, and make complex design decisions are as valuable as they've ever been. The demand isn't going away, it's shifting upward. The same is true for cybersecurity professionals, where the threat landscape is evolving so fast that human expertise, intuition, and real-time judgment are irreplaceable.

The tech industry isn't being hollowed out, it's being restructured. Fewer generalists, more specialists who can do what AI can't.

IT support is a more mixed story. Tier-1 helpdesk work, resetting passwords, walking someone through a standard fix, answering the same five questions for the hundredth time, is exactly the kind of task AI chatbots are eating up right now. But the engineer who crawls under the server room floor at midnight, diagnoses a physical infrastructure failure, and improvises a solution? That person is still very much needed.

Manufacturing tells a similar tale. Automation on factory floors isn't new, robots have been welding car bodies since the 1980s. What's changed is the speed and scope. AI-driven robotics can now handle tasks that once required human dexterity and judgment: quality inspection, precision assembly, even adaptive responses to unexpected variations in materials. Jobs in traditional assembly are genuinely at risk, particularly the repetitive, low-variance ones.

But manufacturing isn't just robots on a line. Process engineers, maintenance technicians, automation specialists, and people who can program, calibrate and repair the machines, their value is actually going up. Every new robot that replaces a human worker needs someone to keep it running. The floor is changing; it isn't emptying.

So What Should You Actually Do?

If you're reading this and feeling a low-grade anxiety, that's reasonable. But anxiety is only useful if it moves you somewhere.

The clearest career insurance right now is to tie your work to things AI fundamentally struggles with: physical presence, emotional intelligence, ethical judgment, creative taste, and human accountability. Not because AI can never develop these things, but because society isn't ready to hand them over yet, and may never fully be.

And if your current role is more on the vulnerable side? The move isn't to panic. It's to consciously reposition. Learn the AI tools in your field before your employer needs to explain them to you. Understand which parts of your job require a human touch and lean into those harder. The people who thrive in this shift won't be the ones who avoided AI, or the ones who were replaced by it. They'll be the ones who figured out how to work alongside it better than anyone else.

How We're Thinking About This at Cubet

Cubet is a software and product engineering company. We build things for clients, web and mobile applications, data platforms, AI solutions. And like every team in this space, we've had to look at ourselves in the mirror and ask the same question we've been exploring in this article: what changes, and what stays?

The honest answer is: a lot changes. And we think that's a good thing.

Over the past couple of years, we've been deliberately weaving AI into how we work, not as a gimmick, and not out of pressure to seem current, but because it makes our engineers sharper, our products better, and our clients' problems easier to solve. We call this our AI³ Approach: three things working together: Intelligence, Innovation, and Impact.

Intelligence means we're applying AI thoughtfully across the entire product engineering lifecycle, from the very first conversation about what to build, all the way through to how we scale and maintain it. It's not a bolt-on at the end of a project. It's baked into how we think from day one.

Innovation means we're constantly asking what's possible now that wasn't possible before. AI has genuinely unlocked new ways to approach architecture decisions, code review, testing, and data analysis. Our engineers aren't spending their best hours on the parts of the job that machines can handle; they're spending them on the parts that actually need a human mind.

And Impact means none of this is for its own sake. Every tool we adopt, every process we evolve, has to show up in the quality of what we deliver. Faster. More reliable. More cost-effective and thoughtfully built.

We're not replacing our teams with AI. We're building teams that are genuinely better because of it.

That means our engineers are learning to work alongside AI tools the way a skilled craftsperson learns to work with better equipment, the craft doesn't disappear, the standard just goes up. Our data and AI specialists aren't being automated out; they're the ones doing the automating. And the people who connect strategy to execution, who understand clients, who ask the right questions, they've never been more central to what we do.

We don't have all the answers. No one does. But we've made a deliberate choice to evolve with this shift rather than wait for it to arrive. If you're working with us, or considering it, that's the kind of team you're getting.

The future of work isn't something that's happening to people. It's something being shaped by the choices companies and individuals make right now. At Cubet, we've made ours. The question is, what is yours?