Your best people aren't quitting. They're looking sideways.

The numbers this year are ugly, but the story under them is the useful part. Your team is using AI every day and hating what it's doing to the work. The job itself is changing shape faster than anyone can teach it. And the people you most want to keep aren't handing in notice. They're drifting.

Let me catch you up on what actually shifted and what it means for how you grow and hold onto talent.

The gap between using it and trusting it

Here's the split that should stop you. A Creative Boom survey of 882 professionals found 86 percent now use AI tools in their work. Only 10 percent think its effect on the industry is positive. That's near-universal use sitting next to near-universal unease.

Don't read the adoption number as buy-in. People are using AI because they feel they have to, not because they're sold. When your team demos a new tool cheerfully, that's not the same as trust. Read it as bracing for change, not embracing it.

The burnout under all this is real: 69 percent hit it in the past year, and mid-career folks lead at 77 percent. Those are the people running your projects and managing your juniors without the power to say no.

Maker to mender

The work landing on desks isn't a blank page anymore. It's something half-generated that needs tidying. One anonymous designer put it plainly in a Creative Boom advice column: "I used to feel like a maker. Now, I feel like a finisher. A correction service."

That's not just a mood. It's a skills problem. If people stop starting from scratch, the muscle for it goes soft. Claire McDivitt at Lazerian framed the risk well: the danger isn't the new tools, it's forgetting to make time to create something from a blank page.

Some people are already voting with their feet. Others are pushing back on clients. But the quiet cost is the one you can't see on a dashboard: your senior people getting slower at the thinking that made them worth hiring.

The drift you won't spot until it's late

Here's the retention trap. Only 7.5 percent plan to leave the field entirely. That sounds fine. It isn't. The real problem is people staying in the profession while looking sideways, out of your team, your agency, your setup.

A wave of resignations is visible. A slow leak of disengaged, undervalued people is not, until they're gone. Half of respondents feel less financially secure than a year ago. More than a third are considering a job change. Add the maker-to-mender ache on top, and you've got talented people mentally checked out while still on payroll.

So what do they actually want? Not more software. Asked what would improve their working lives, 57.5 percent said networking and community, 53 percent said mentorship. New tools trailed at 31 percent. They want people, not plug-ins.

When you can't out-experience the problem

Mentorship is the obvious answer, but there's a catch nobody warns you about. A designer with 20 years of experience may have only two or three years in an AI-first world. As one professor put it in When the profession outruns the mentor, expertise now has "something resembling a half-life."

That changes what your senior people should teach. Not execution, since AI does that. The job now is judgment: telling a junior why option four's hierarchy fails, why the spacing fights the content. "A model can generate six interface concepts before a junior designer finishes a sketch," he writes. What it can't do is explain why one is weak. That explanation is the mentorship now.

This lines up with the reframe from Cannes Lions 2026, where Jon Williams of The Liberty Guild argued AI isn't changing creativity, it's changing where value sits. Creative judgment, he said, is becoming the most valuable commodity, and it's "pretty damn hard to prompt."

The deep cut

The layoffs-for-AI bet is already unwinding, and that's your opening. Companies that cut people in 2025 for the almighty AI tools now regret it, as one UX Collective piece notes, because token costs versus a full salary don't pencil out. Designer Andy Strong made the same call: subsidized subscriptions are giving way to pay-per-generation, and the true cost is coming.

So don't budget as if execution is now free. Budget for the thing that's getting scarce: people who can look at generated output and say precisely why it's wrong. Protect your mid-career people from burning out on cleanup, because they're the ones who carry that judgment and teach it. Give juniors real blank-page work on purpose, even small internal projects, so the muscle doesn't atrophy before they're senior enough to mentor. The move on Monday isn't a new tool. It's carving out time for the two skills AI can't hand you: starting from nothing, and knowing why something's good.

Three questions for your team

1. Who on your team has drifted from maker to mender, and what real blank-page work can you hand them this month to pull them back?
2. When your seniors mentor, are they teaching execution AI already covers, or the judgment to say why an answer is weak? Which one are you rewarding in reviews?
3. If token pricing keeps climbing, does your next-quarter plan still assume design labor is cheap, or are you funding the human judgment you'll need when the AI math stops working?