One of the biggest questions raised at our recent Round Table was about apprenticeship, and what happens to it once AI takes over the parts of design work that used to feel like the boring bits.
Transcript analysis, synthesis, exploration, repetitive production: none of it looked like the interesting part of the job. It looked like the part you got through before the real thinking started.
The uncomfortable question the room sat with is whether that “boring” work was the training ground where judgement got built. The follow-on question is what happens to the next generation of designers if it disappears before they have had a chance to learn from it.
What AI has removed from junior design work
Think about what those tasks involved.
Reading through hours of interview transcripts to pull out themes, and sitting with a research finding long enough to see the pattern underneath it, was one part of the job. Producing version after version of a screen or a flow until one of them finally felt right was another.
None of that was throughput for its own sake. Each pass was a small decision: is this the right theme, does this pattern hold up, does this version solve the problem better than the last one.
A designer who did that work hundreds of times over a few years was building more than output. They were building a private library of small calls, most of them wrong at first, that eventually turned into instinct.
AI can now do the transcript analysis, the synthesis, the exploration and a good share of the production in a fraction of the time. That is a real gain for anyone who already has the instinct to check the output against.
It is a different story for anyone who does not have that instinct yet, because the removal of the work also removes the place where that instinct used to get built.
Why seniors move fast with AI and juniors cannot yet
Here is the pattern I keep seeing. Senior designers can pick up AI tools and accelerate almost immediately, because they already carry years of prior experience telling them when an answer does not feel right.
They can look at an AI-generated set of themes, or a first-pass prototype, and know within seconds whether it holds up. That judgement did not come from nowhere.
It came from all the slower, more repetitive work they did earlier in their careers, the exact kind of work AI now removes.
A junior designer using the same tools does not have that internal check yet. They can produce a plausible-looking output as fast as a senior can, but they have no reliable way of knowing whether it is any good.
The tools have gotten faster than the people using them have had the chance to build the judgement that makes speed safe, and that gap will not close on its own as the tools keep improving.
Speed without judgement is not faster, it is a plausible-looking answer arriving sooner, with no one yet able to tell whether it holds up.
That is exactly why it is worth being deliberate about what junior designers still do without AI, so the next generation gets to build the same instinct experienced designers already carry, rather than skipping past the work that used to teach it.
How to protect junior judgement while using AI
This is where it stops being a philosophical question about the future of design and becomes a decision you are already making, whether you have noticed it or not.
Every time you approve a leaner team, or decide a junior hire is not needed because AI covers that now, you are making a bet.
The same goes for letting a junior work independently with AI tools from their first week: you are betting on where the next generation of senior judgement is going to come from.
If juniors spend their early years directing AI output rather than sitting inside the slower work that used to build instinct, that instinct has to come from somewhere else. It is worth being honest that nobody has a proven answer for where yet.
The practical response is to be deliberate about what juniors are still exposed to, so the tools speed up their output without skipping the part of the job that used to build their judgement.
Keep some unassisted, judgement-testing work in junior design briefs, and ask a junior to make a call on a research theme or a design direction before they check it against AI output.
Build in review moments where a junior has to explain why an answer is right or wrong, instead of presenting the output alone. That single habit, repeated often enough, does some of the same work the old repetitive tasks used to do.
None of this slows the team down in any meaningful way. It makes sure the speed you are getting from AI today does not come at the cost of the judgement you will need from that same designer in five years.
Ready to talk about your next design hire? Get in touch.
Liam
Design, Creative & Research Recruitment
Last updated: September 2026
