Which of Your Strengths Will Become More Valuable as AI Improves?
You've read the lists. Ten skills for the age of AI. Five things robots supposedly can't do. Somewhere on the list is a strength you already have, and for a moment that feels reassuring, until you notice the list says nothing about your version of that strength, in your particular role, doing your particular kind of work. It can't. A list built to apply to everyone ends up applying, in any useful sense, to no one.
Underneath most of that advice sits a simpler instruction: play to your strengths, get even better at what you already do well. It sounds unimpeachable. It's also missing a test. Not every strength is worth developing further, and the instinct to assume otherwise can be an expensive one.
Two financial analysts make the problem concrete. Both were strong at their jobs. One built her reputation on detailed, accurate, beautifully constructed models. Stakeholders trusted her numbers, and for three years that trust protected her position. Then AI-powered modelling tools arrived. A colleague generated a revenue forecast in forty minutes that would have taken her two days, and the quality was acceptable enough that leadership used it without complaint. She told herself the tool was fine for simple work but that anything complex still needed her. Six months later, requests for her modelling work had fallen by nearly a third. Her skill hadn't declined. The thing she was skilled at had become cheap to produce another way.
This is the trap in "double down on your strengths." It treats every strength as equally durable, when what actually determines durability isn't how good you are at something. It's what kind of thing it is.
A test that has nothing to do with how good you are
The useful question isn't whether you're strong at a task. It's whether that strength multiplies when AI gets involved, or simply gets absorbed by it. A strength that primarily helps you finish something faster offers limited protection, because speed is exactly what improving tools are best at closing the gap on. Automation doesn't distinguish between a task done in thirty minutes and one done in three hours; it cares whether the task itself is procedural enough to learn. A strength that helps you frame a problem correctly, resolve genuine ambiguity or spot a risk before it becomes expensive behaves differently. As AI absorbs more of the surrounding execution, that kind of judgement doesn't get replaced. It becomes the part of the work that's left, which means it becomes more central, not less.
Two questions do most of the work here. First: when you use this strength well, does it mainly change how quickly something gets done, or does it change the decision itself, the way a stakeholder acts, the risk that gets caught in time? Second: if AI took over everything routine around this strength tomorrow, would your involvement become more essential to what remains, or would there be less for you to do? A strength that survives both questions is worth investing in. One that only survives the first is worth keeping at its current level and no more, because the return on developing it further is already thinning.
This is really the same idea the wider question of irreplaceability rests on. What makes someone hard to replace was never about effort, tenure or being first in and last out. It's about consequence: what would degrade if this person's involvement disappeared, and how expensive would that degradation be to absorb? Speed can be automated regardless of how fast you already are. Significance is harder to automate, because it depends on judgement that's specific to context, not on the mechanical completion of a task.
Applying the filter to your own work
A product manager who went through this filtering exercise identified three strengths worth developing further: turning scattered user feedback into a clear set of insights, identifying the trade-offs between what users wanted and what the platform could actually support, and explaining product decisions in a way that reduced confusion across teams that didn't always talk to each other. She stopped putting effort into roadmap maintenance and user-story writing and let AI handle both instead. Within six months her calendar had visibly changed shape. She stopped attending the weekly roadmap update meeting, because the roadmap updated itself. She started joining strategy conversations two weeks earlier than before, because leadership wanted her synthesis of user feedback before direction got set, not after. She was attending fewer meetings, but the ones she attended mattered more. The filtering hadn't just clarified what to develop. It had changed how people around her used her.
Not every strength sorts neatly into "invest more" or "let it go." Some sit in a genuinely neutral category, where AI neither multiplies them nor threatens to replace them. Relationship management tends to fall here: nothing about improving AI tools deepens your ability to build trust with a colleague, but nothing about them erodes it either. The sensible response to a neutral strength is to maintain it at a reasonable standard and stop there, because further investment produces a flat return rather than growing leverage.
If you want to try this on your own work, pick two or three things you're currently strong at, ideally ones that take up real time. For each, ask honestly whether the value you create comes from finishing the task or from what your judgement does to the outcome once it's finished. Notice which ones would keep mattering, or matter more, if the routine parts of the job around them were handled elsewhere. Those are the strengths worth your development time. The others may still be necessary, competent work, without being the thing your career should be built on.
It's worth adding what this filter isn't. It isn't a claim that any human capability is permanently beyond the reach of AI, and it isn't a reason to treat today's answer as fixed. Capability moves in increments rather than announcements, and a strength that scores well now can drift towards commodity later as the tools around it improve further. The filter is something to apply periodically, not once and forget.
What it does give you is a way to stop guessing. The strength worth developing is rarely the one you enjoy most, or execute fastest, or feel most identified with. It's the one that grows more consequential precisely as everything around it gets easier to produce. That's a different question from "what am I good at," and it's the one that actually determines where your effort should go next.
David Taylor
Interested in going further?
The AI-Ready Career builds on this same filtering logic to show how professionals redesign entire deliverables, not just individual strengths, around the judgement that AI can't absorb.
Ad · Amazon affiliate link.
