And now you find yourself reading about large language models that can produce a first draft of almost anything in thirty seconds. You sit in meetings where junior colleagues talk about AI as though it is simply a faster version of you. You wonder, privately, whether the thing you spent all those years building is about to become less useful. Less valued. Less necessary.

This is not imposter syndrome. It is a specific, legitimate concern — and it is affecting senior professionals more acutely than anyone in your organisation is likely to acknowledge.

Before we look at what the evidence actually suggests, it is worth pausing on why this particular anxiety tends to run quietly rather than loudly.

When AI disruption is discussed publicly, the conversation tends to focus on entry-level job displacement — call centres, data entry, routine administrative tasks. That framing, while accurate in some respects, misses the subtler anxiety that sits in senior roles.

The fear at Director or Senior Manager level is not "will I lose my job tomorrow?" It is something harder to articulate: "Will the thing that makes me valuable — my accumulated knowledge, my contextual judgement, my professional reputation — gradually become a commodity that any reasonably capable AI can approximate?"

That is a different kind of worry. And it deserves a more precise answer than most AI commentary provides.

The short version: the evidence does not support the worst-case narrative — but it does require a specific kind of adaptation.

The World Economic Forum's Future of Jobs Report 2025 identified the skills expected to grow most in value over the coming years. The list is notable for what it includes: analytical thinking, creative thinking, resilience, leadership, and the ability to work effectively with others in conditions of uncertainty. These are not skills that AI replicates well. They are, largely, the skills that accumulate with experience.

The report also found that the roles most vulnerable to displacement are those built around single, repeatable tasks — not roles that require integrating multiple domains of knowledge, managing relationships, or exercising judgement in ambiguous situations. Senior professional roles, in most sectors, are predominantly the latter.

This aligns with research published in Daedalus by economist Erik Brynjolfsson, who draws a distinction between AI designed to replace human workers and AI designed to augment them. His argument — supported by productivity data — is that augmentation produces better outcomes for both organisations and workers, particularly those whose value lies in complex problem-solving rather than task execution. Experienced professionals, in Brynjolfsson's analysis, benefit disproportionately from AI that amplifies their capacity rather than competing with it.

McKinsey's 2023 analysis of the economic potential of generative AI reached a similar conclusion: knowledge workers who learn to work with AI tools see productivity gains; those who treat AI as a threat and disengage see relative decline. The differentiating factor is not age, sector, or seniority level. It is whether someone learns to direct the technology rather than compete with it.

None of this means the transition is frictionless. But the headline finding is worth sitting with: experience does not become irrelevant. The question is whether it is applied in ways that make the most of what AI cannot do.

Most of the anxiety around AI and experience is actually several different concerns bundled together — and bundled concerns are hard to act on.

There is the concern about your organisation's specific decisions and how AI adoption might affect your role. There is the broader question of how your industry is changing and whether your skills map onto where it is heading. There is the identity question: if AI can produce a version of your output, does that change what your output means? And there is the practical question of whether you need to learn new tools, and if so, which ones.

These are distinct problems. They require different responses. And some of them — particularly the ones about your organisation's decisions or the pace of industry change — are not within your control to resolve today, regardless of how much mental energy you spend on them.

The Sorting Method is a simple framework for untangling anxiety — it asks you to sort your concerns into three categories:

Most AI anxiety, when you sort it carefully, concentrates in the third column. That does not make the worry irrational — the concerns are real. It simply means that spending mental energy on them today does not change the outcome. What does change outcomes is clear action in the first column.

One concern that often sits beneath the surface of this conversation is harder to sort: what does it mean for your professional identity if AI can replicate aspects of your work?

This is worth its own attention — and it is the subject of a separate article in this series. But briefly: the value of your experience is not located in your outputs alone. It lives in your judgement about which outputs matter, your capacity to ask the right question rather than answer the wrong one efficiently, and your ability to earn trust in contexts where trust is not automatically granted.

AI does not replicate those things. It can produce a version of your deliverable. It cannot produce your perspective on whether that deliverable is the right response to the actual situation.

For a deeper exploration of this theme, see: Professional Identity in the Age of AI: The Question Nobody Is Asking

If you have arrived at this article because you are genuinely worried about your relevance in the AI era, the most useful next step is not a course, a certificate, or a reading list.

It is this: spend fifteen minutes this week sorting your specific concerns using the framework above. Write them down. Separate what you can act on from what you cannot. Then take one action — the smallest one that generates real information.

The Calm Action Cycle — the framework that underlies this approach — follows the same logic: Sort, See, Try. Each step is small enough to attempt without certainty. Each attempt generates something useful, regardless of outcome.

Read more:

• AI Anxiety at Work: A Calm Guide to Career Resilience in the Age of Disruption
• Nighttime Anxiety & 3am Overthinking: The Complete Guide
• Life Transitions: The Complete Guide to Navigating Change with Grace
• Career Change at 40: Is It Too Late to Start Over?