The senior mindset and why you need it in the AI age

By Zaynah Danquah and Justice Jones Zaynah Danquah is lead instructional designer and a co-founder of 24/7 Teach. Justice Jones is co-founder and Chief Strategy Officer. Published August 12, 2026. Reading time: 10 minutes.

Here is the advice going around: stop waiting for direction, take ownership, think strategically, act like a senior even if your title says otherwise.

All of it is correct. None of it answers the question underneath, which is where senior judgment was supposed to come from in the first place.

It came from years of doing simpler work under supervision. You took the small tasks, you got them wrong in a setting where wrong was survivable, someone with more experience told you why, and slowly the pattern library assembled itself. Nobody taught you judgment in a session. You accumulated it by being trusted with things that did not matter much, until you could be trusted with things that did.

That rung is disappearing. And telling people to arrive already senior, without saying anything about how, is a door that looks like an invitation.

What actually changed

Three economists at Stanford, Erik Brynjolfsson, Bharat Chandar, and Ruyu Chen, went looking for evidence in payroll data rather than in predictions. Using ADP administrative records covering millions of workers, their November 2025 revision reported that early-career workers aged 22 to 25 in the most AI-exposed occupations saw a 16% relative decline in employment, controlling for firm-level shocks, while employment for more experienced workers in the same occupations stayed stable.

Two details matter more than the headline.

First, the declines concentrated in occupations where AI automates work rather than augmenting it. Where the technology helped people do their jobs, employment held. Where it did the job, entry-level hiring fell. That is a much more specific finding than "AI is taking jobs," and it points at which rungs went missing.

Second, this is a working paper and the causal question is genuinely contested. The post-2022 hiring pullback started before generative AI was widespread, and interest rates are a competing explanation. The same authors published a follow-up in February 2026 arguing that rate changes do not account for the concentration in AI-exposed occupations specifically. That is a reasonable answer, not a closed case. If you see this cited as settled fact, the person citing it has not read the follow-up. You will also see the number reported as 13%, which was the figure in the earlier draft before the November revision.

What is not in dispute is the mechanism, and the mechanism is the part that should worry you. Entry-level work was never only work. It was the training subsidy. Firms could afford to teach because juniors were cheap relative to what they produced, and the gap between the two got repaid later. When a tooled-up experienced person can produce the same output, that arithmetic stops working, and firms stop assigning the tasks through which people used to develop.

So the loss is not just the job. It is the learning that used to be attached to the job.

Why this lands hardest on people who did everything right

The people most exposed here are the ones who followed the instructions. Get the credential, get the entry-level role, work your way up. The credential still gets issued. The rung it was supposed to reach has moved.

That includes career-changers most of all. If you are thirty-four and moving from teaching into learning design, or from accounting into operations, or from nursing into health tech, you were counting on the same ladder a twenty-two-year-old was counting on. You have more life experience than they do, but you have the same amount of experience in the new field, which is none, and you are competing for the rung that got removed.

And there is a second trap sitting right behind the first, which is that AI makes it easy to produce work that looks like it came from someone who knows what they are doing.

Researchers at Microsoft Research and Carnegie Mellon surveyed 319 knowledge workers who use generative AI at least weekly and collected 936 first-hand examples. They found something uncomfortable: confidence in the AI predicted less critical thinking, while confidence in oneself predicted more. The people least equipped to catch a flawed output were the least likely to look for one, because a fluent draft gives a novice no reason to be suspicious.

Put those two findings next to each other and the shape of the problem is clear. The work that used to build judgment is being handed to a tool, and the tool produces output that a person without judgment cannot evaluate. That is not a motivation problem. It is a structural one, and it is worth being honest that no individual reading this article caused it.

The thing that has to replace the apprenticeship

Here is where it turns, and the turn comes from adult learning research rather than from career advice.

Malcolm Knowles, who spent a career on how adults learn differently from children, built his framework on a first assumption that turns out to be exactly load-bearing here: as a person matures, their self-concept moves from dependent toward self-directed. Adults can identify what they need to learn, choose how to pursue it, and evaluate whether it worked. He later wrote a whole book on self-directed learning as a practice rather than a trait.

Knowles's framework is debated, including whether the split between how adults and children learn is as clean as he claimed. But the self-direction assumption is the best-supported part of it, and it is the part that matters, because the apprenticeship was a system that directed your development for you. Someone chose your tasks, sequenced your difficulty, and told you when you were wrong. With that system thinning out, the direction has to come from somewhere, and the only remaining candidate is you.

This is the real content of a senior mindset, and it is not attitude. It is running your own apprenticeship: choosing work that stretches you slightly past competence, seeking the correction nobody is obligated to give you, and building the pattern library deliberately instead of absorbing it by accident.

Which means the practical question is not "how do I seem more senior." It is "what did the apprenticeship do for me, and how do I do that for myself."

The four-step move that does most of the work

The apprenticeship's core loop was: attempt something, have someone with more experience react to it, adjust. Nearly all the learning was in the second step. So the single highest-leverage habit is the one that generates reactions.

It comes from our own mentoring practice, and it has four steps. When you hit something unclear:

One. Name the uncertainty out loud. Not to yourself. In writing, to someone. "I am not sure whether this should cover X or stop at Y." Naming it is what makes it available for correction.

Two. Work the problem before you take it to anyone. Decide what you think should happen and why. This step is uncomfortable and it is where the learning actually lives, because guessing and finding out you were wrong builds judgment faster than being told the answer.

Three. Turn your thinking into something concrete. A proposal, an outline, a draft, two options with a recommendation. Something a person can react to specifically rather than generally.

Four. Bring it for alignment, not for permission. "Here is what I think we should do and why. Does that match how you see it?" You are asking someone to check your reasoning, not to hand you the answer.

The test is simple and it is not subjective: did you arrive with a proposal, or did you arrive with a question? A question gets you an answer, which solves today and teaches nothing. A proposal gets your reasoning corrected, which is the thing you were supposed to get from three years of supervised work.

There is a version of this that fails, and it is worth naming. Bringing a confident proposal you have not actually thought through is worse than asking a question, because it wastes someone's attention and trains them to stop reading your work carefully. Step two is not optional.

Do this twenty times and you have had twenty rounds of judgment correction. That is not a substitute for an apprenticeship. It is the closest available thing, and it is available now.

Write down why, not just what

The second habit is smaller and compounds harder: document the reasoning behind your decisions as you make them.

Not the decision. The reasoning. What you were optimizing for, what you gave up, what you were unsure about, what would have changed your mind.

This does three things at once. It forces the reasoning to exist, because vague thinking survives in your head and dies on the page. It gives you a record to check against outcomes later, which is how you learn whether your judgment is any good rather than just feeling confident about it. And when someone asks why you did it that way, in a review or an interview or a meeting where your standing is on the line, you have an answer that is specific.

That last part is increasingly the whole game. When output is cheap and anyone can produce something polished, the differentiator is whether you can explain what you did and defend it under questioning. A portfolio shows what you made. The reasoning shows whether you knew what you were doing, and it is the part that cannot be generated for you.

The objection: this assumes it is safe to propose

Everything above assumes you can bring a proposal without it costing you. For a lot of people, that assumption is wrong, and pretending otherwise would make this article useless to exactly the people it is for.

If you are new, on a contract, or on a team where initiative reads as overstepping, "bring a proposal instead of a question" is advice from someone whose standing makes proposing free. There is also a version of this that has nothing to do with the workplace at all. Some people were raised to defer to elders and to authority, and treating that as a professional deficit is both wrong and useless. It is not a character flaw. It is a real thing to work through, and it is worth naming rather than pretending everyone starts from the same place.

Three honest responses.

The format is adjustable. "Here is what I think, does that match how you see it" can be delivered as a firm recommendation or as a genuine question, and the learning is identical either way. What matters is that your reasoning becomes visible and gets corrected. The confidence level is a delivery choice, not the substance. Match it to your actual standing rather than to how a career article says you should sound.

You can lower the stakes. Practice on small, reversible decisions first, where being wrong costs nothing and nobody is watching closely. Build the habit where it is cheap, then use it where it counts.

Sometimes the environment is the problem. If proposing genuinely is not safe where you are, no mindset shift fixes that, and reframing a structural constraint as a personal one is a way of blaming people for their circumstances. That is a job problem wearing a skills problem's clothes. Worth diagnosing correctly, because the fix is different.

And the larger caveat holds for the whole article. If firms stopped hiring juniors, the fix is firms resuming it, or programs and apprenticeships deliberately rebuilding the rung. Individual habits do not solve a structural problem. They are what is available to you while the structural problem gets worked out by people with more leverage than either of us.

What to do this month

1. Pick one thing you are currently unclear about and run the four steps on it. Name the uncertainty in writing, work the problem, build a proposal, bring it for alignment. One round this week. The habit forms through reps, not through agreeing with the idea.
2. Start a decision log. Date, decision, what you were optimizing for, what you gave up, what you were unsure about. Ten entries in and you will start seeing your own patterns, which is what a mentor used to do for you.
3. Find one person who will tell you when you are wrong. Not a cheerleader. Someone whose correction you would actually take. This is the single hardest thing to replace about supervised work, and it does not happen unless you ask for it explicitly.
4. Check your own drafts before you trust them. If AI is producing work you cannot yet evaluate, that gap is your development plan. Pick the part you cannot check and learn that first.
5. Choose one task slightly past your competence, on purpose. The apprenticeship worked by sequencing difficulty. Nobody is sequencing it for you now, so sequence it yourself, and pick something where failing is survivable.

The advice to take ownership and think strategically was never wrong. It was just incomplete, because it described the output of an apprenticeship without mentioning that the apprenticeship was disappearing.

Judgment still comes from attempting things and being corrected. That part has not changed and will not. What changed is that nobody is going to arrange it for you.

Related reading

• The solutionist mindset needs a stopping rule — when to stop adapting to a constraint and name it instead.
• The minimum in minimum viable analysis is not a fixed number — how much thinking a decision actually deserves.

Career services at 24/7 Teach

24/7 Teach builds programs around real projects and real client work rather than practice exercises, because judgment comes from attempting things and being corrected. Career services include coaching through job transitions and consulting on high-level performance to secure placement or readiness for promotion once hired. More than 600 adults have been placed in new careers through our programs. See programs and upcoming cohorts

About the authors

Zaynah Danquah is the lead instructional designer and a co-founder of 24/7 Teach, where she designs programs and curriculum across the company's teen, adult, and organizational tracks. Full bio →

Justice Jones is co-founder and Chief Strategy Officer of 24/7 Teach and a former K-12 principal. Through 24/7 Teach, he and his team have supported more than 50 organizations and placed more than 600 adults in new careers. Full bio →

This article was researched and written by Zaynah Danquah and Justice Jones with AI assistance, then reviewed and edited by our team. External studies and sources are credited to their original authors. Examples from our own work reflect our organizational practice.