I have been in the corporate world for 28 years. I have watched technology disrupt industries several times over. I have seen entire job categories disappear. I have watched people who built careers on specific technical skills suddenly find those skills commoditised overnight. Each disruption followed a similar pattern: the people who survived and grew were not the ones who tried to out-compete the new technology. They were the ones who moved to a level the technology could not reach.
What is happening with AI today is that pattern playing out at a speed and scale I have not seen before. The difference this time is that AI is not just automating tasks. It is generating thinking. And that changes the conversation about what human beings need to bring to the table to remain not just employed, but genuinely valuable.
Over the next five months, I want to have that conversation properly. Not with buzzwords and vague reassurances, but with the specific, practical honesty that I wish someone had offered me earlier in my career. This series is about the skills that are becoming more valuable in an AI-heavy economy, not less. And the reason each of these skills matters is the same: they sit in territory that AI cannot reliably own from end to end.
What Is Shifting in the Economy
When most people talk about AI taking jobs, they frame it as a threat. That framing is understandable but incomplete. A more accurate picture is this: the nature of valuable work is shifting. Work that is repetitive, pattern-based, and well-defined is migrating rapidly to AI systems. Work that requires judgment, context, trust, and human connection is becoming more scarce, and therefore more valuable.
Think of it this way. For most of the last century, organisations needed large numbers of people to execute defined tasks. The skills that got you hired were the ones that made you reliable and consistent. Today, organisations have access to systems that are more reliable and more consistent than any human being. What they still need humans for is something different entirely: they need people who can direct those systems wisely, make sense of ambiguous situations, build trust with other humans, and take responsibility for decisions that carry real consequences.
The highest-value skills are shifting from doing repetitive work to directing systems, making decisions, and creating leverage. This is not a small change. It is a reordering of what professional competence means.
This Series: 10 Skills AI Cannot Replace
July 2026
August 2026
September 2026
Part 1: Problem Framing and Critical Thinking
AI is genuinely impressive at generating answers. Give it a well-formed question and access to the right data, and it will produce outputs that would have taken a team of analysts days to compile. That capability is real and growing quickly. But here is the thing most conversations about AI miss: generating answers is only half of what valuable work actually requires. The other half is knowing which question was worth asking in the first place.
This is where human judgement remains not just relevant but irreplaceable. Defining what the real problem is, separating the important from the merely urgent, knowing when the question you are being handed is the wrong one entirely, these are acts of critical thinking that require context, experience, and something that cannot be extracted from data alone: wisdom about what actually matters.
"AI is strong at generating answers. Humans are still needed to ask the right questions. And in organisations where everyone has access to the same AI tools, the person who asks better questions is the person who wins."
What Problem Framing Actually Means
Problem framing is the act of defining what you are actually trying to solve before you start solving anything. It sounds obvious. In practice, most organisations skip it entirely.
Most meetings, projects, and decisions in the corporate world start at the solution stage. Someone brings a problem-shaped statement that already contains an assumed answer, and the group spends its energy debating execution details rather than asking whether the premise is correct. This happens because defining a problem properly is uncomfortable. It requires sitting with uncertainty. It requires being willing to say "I am not sure this is the right question yet" in a room full of people who want to move fast.
Problem framing as a skill involves several things working together.
Separating the presenting problem from the real problem
The problem that arrives on your desk is almost never the root problem. A team experiencing communication breakdowns is probably experiencing a leadership or clarity problem upstream. A product that is not selling is probably experiencing a positioning or market fit problem, not a feature problem. The ability to look past the surface description and identify what is actually happening is where real value begins.
Challenging the assumptions underneath a question
Every question someone brings to you comes loaded with assumptions they have already made. A question like "which of these three vendors should we choose?" assumes that choosing a vendor is the right next step. A question like "how do we improve our sales conversion rate?" assumes that conversion rate is the lever that matters. Critical thinkers surface these assumptions before committing to an answer. They ask: what has to be true for this to be the right question?
Making decisions with incomplete information
One of the most important things I learned in 28 years of corporate life is that waiting for complete information before making a decision is not wisdom. It is avoidance disguised as rigour. Real decisions are almost always made under uncertainty. The skill is in knowing which missing information is material enough to wait for, and which missing information you can work around. That judgement is not something AI can reliably make, because it requires understanding of the specific stakes, relationships, and consequences at play.
Thinking strategically instead of mechanically
Mechanical thinking follows a process. Strategic thinking asks whether this process is the right one given this particular situation and these particular goals. It is the difference between a manager who executes a plan well and a leader who knows when to change the plan. AI can optimise within a defined framework. It cannot yet reliably tell you when the framework itself needs to change.
A Real Example from the Corporate World
A company wants to launch a marketing campaign for a new product
The team feeds their brief into an AI tool. Within minutes, they have a fully developed campaign: messaging, audience segments, channel mix, budget allocation, creative concepts, a content calendar, and projected reach numbers. It is thorough, well-structured, and delivered faster than any agency ever has.
But before any of that output becomes useful, a human being has to answer several questions that the AI cannot answer for them.
A complete campaign for the market segment specified in the brief, with optimised messaging and channel recommendations based on available data.
Whether this market segment is actually the right one to target. Whether the customer pain the campaign is addressing is real or assumed. Whether this campaign aligns with the business strategy, or whether it is the right quarter to spend on acquisition at all.
The AI produced the campaign excellently. But the human judgement that decides whether that campaign should exist at all, whether it addresses the real problem, and whether it serves the actual business goal: that judgement is where the value sits. And that judgement requires critical thinking that no AI system currently owns end to end.
Why AI Cannot Fully Own This Skill
AI systems are trained on historical data and optimised for pattern recognition. They are genuinely powerful within defined domains. But problem framing requires something different: it requires the ability to question the domain itself.
When an AI is given a question, it answers that question. It does not step back and ask whether the question is worth answering. It does not bring a decade of experience working in a specific industry to bear on whether this particular situation is actually as described. It does not feel the political weight of a decision, or understand that the real agenda in this meeting is not what anyone has said aloud. These are human capacities built from experience, observation, and the ability to hold complexity and contradiction simultaneously.
I want to be honest here: AI is getting better at some of this. But "better at some of this, in specific conditions, with well-structured inputs" is very different from "reliably owns this end to end in the messy, ambiguous reality of actual organisations." We are not close to that. And the organisations I see thriving are not the ones waiting for AI to get there. They are the ones building teams where human critical thinking directs AI capability rather than competing with it.
"The professionals who will be disproportionately valuable in the next decade are not the ones who use AI. Everyone will use AI. The valuable ones are the ones who know what to ask, when to push back on the output, and how to frame the problem that AI is being pointed at."
How to Build This Skill Deliberately
Problem framing and critical thinking are not fixed traits. They are capacities that develop with deliberate practice. Here is what that practice looks like in a professional context.
Not describing the situation. Defining the specific problem you are trying to solve. This one habit alone separates good thinkers from excellent ones. You will be surprised how often the act of writing it down reveals that you were solving the wrong thing.
Every plan rests on assumptions. Most people never name them. When you name the three assumptions your approach depends on, you can test them before you commit. You often find that one of them is not as solid as you thought.
Not out loud every time. But internally, as a discipline. The presenting topic in most meetings is a surface level. Underneath it is usually something about priorities, resources, relationships, or strategy. Training yourself to see that layer is what critical thinking looks like in practice.
The reflex to resolve uncertainty quickly is natural and almost always counterproductive. Developing a tolerance for sitting with an open question, gathering multiple perspectives before concluding, and resisting the first plausible answer is what makes critical thinking a real capability rather than just an aspiration.
You cannot develop problem framing skills entirely alone because the blind spots in your thinking are, by definition, the things you cannot see from where you are standing. A skilled coach who asks the right questions, and refuses to let you slide past an assumption you have not examined, accelerates this development faster than any other method I have seen.
What This Means for Your Career
If you are a student choosing a field of study, a young professional trying to differentiate yourself, or a mid-career professional wondering how to stay relevant over the next decade, this is the central message of this series: invest in the skills that sit above the work, not just in the work itself.
Technical skills will continue to be important. But in a world where AI can produce technical outputs at scale, the person who decides what outputs to produce, and why, and for whom, and whether the framing is right in the first place, that person becomes disproportionately valuable. That is not a prediction. That is already happening in the organisations I work with.
The next month of this series will look at how communication and influence work in an AI-driven world, and why the ability to make ideas land with human beings is something no AI can do reliably on your behalf. I hope you will join me for it.
Developing These Skills Takes a Coach
Critical thinking and problem framing are not things you develop by reading about them. They develop through practice, feedback, and someone who holds you to a higher standard of thinking. That is what coaching provides.
Book a Free Conversation