The generation that learned to work without AI holds the one thing the technology cannot supply on its own the judgment to know when the machine is wrong.
By Rajiv | Opinion
Every technology wave arrives with the same background hum of anxiety, but this one has moved faster than most. Professionals in their late thirties, forties and fifties are watching software rewrite the daily texture of work — drafting the memo, reconciling the ledger, summarizing the contract, producing the first version of the code — and quietly wondering whether two decades of hard-won experience have just been marked down.
The worry is understandable. It is also, in most cases, misplaced. What is being automated is not the role; it is the first draft of the role. And a first draft is only valuable to someone who can tell a good one from a bad one.
Start by separating tasks from work
The most useful exercise a mid-career professional can do this quarter takes an afternoon. List everything you did last month, then sort it into three buckets: what a capable model could now do in seconds, what it could do with your supervision, and what still depends entirely on you knowing the organization, the customer or the regulation. Most people are startled by how large the first bucket is and relieved by how much sits in the third.
That third bucket is the career. It is rarely made of technical steps. It is made of context: why this exception was approved last year, which supplier will escalate, what the auditor actually cares about, how a decision will land with the team that has to live with it. None of that is written down anywhere a model can read.
Learn the failure modes, not just the features
Younger colleagues will out-pace you on tool fluency, and that is fine fluency is cheap and getting cheaper. The scarcer skill is knowing where these systems break. They are confident when they should hesitate. They smooth over the awkward detail that mattered. They reproduce the bias in the data they were trained on. They are excellent at plausible and indifferent to true.
Someone who has spent fifteen years watching processes fail in specific, boring, expensive ways is unusually well equipped to spot exactly this. That instinct is not nostalgia. It is quality control, and it is about to become one of the better-paid capabilities in the market.
Rebuild the learning habit before you need it
The real risk for experienced professionals is not that they cannot learn. It is that they stopped practicing learning somewhere around year eight, when competence made it unnecessary. Getting the habit back is uncomfortable and entirely doable: an hour a week, a real problem from your own job, and a tolerance for looking clumsy in front of people who are better at it than you.
Choose depth over breadth. One serious use case, taken from idea to something colleagues actually rely on, teaches more than a dozen certificates. It also produces the thing recruiters and internal sponsors respond to — evidence rather than intention.
Lead the adoption instead of waiting to be trained
Organizations are not short of AI tools. They are short of people who understand a process deeply enough to say where the technology should and should not touch it. That is a Gen X and millennial job description almost by default, and it is far more durable than being the fastest prompt writer in the room.
The instinct to wait for a corporate training programme is the one habit worth breaking now. The people who will do well are simply those who started early, chose their own problem, and were willing to be visibly mid-level at something again.
The advantage nobody talks about
There is a narrow window in which having worked both before and during this shift is itself a qualification. You know what the process looked like when it was entirely human, so you can see precisely what the automation improved and what it quietly broke. The cohort arriving after you will not have that comparison.
Experience is not being devalued by AI. It is being repriced away from execution and towards judgment. For anyone who has spent fifteen or twenty years accumulating exactly that, this is a better bargain than it first appears.
Rajiv Tiwari leads GBS AI Transformation at Technip Energies, is a Six Sigma Master Black Belt and writes on operations, technology and the future of work.
Follow on You Tube – https://www.youtube.com/@rajivtiwari02
