Half Your Team Feels Like a Superhero. The Other Half Is Losing the Plot.

A year ago the mood in tech was "burned out, but optimistic." That was the easy version. The 2026 numbers are messier, and they land right on your desk. Your team is no longer one group with one feeling about the work. It has split in two, and the split runs straight through your roster. Let me catch you up.

One question now predicts more than title or tenure

The old way to guess how someone felt about their job was to ask what they did and where they worked. That's over. In Lenny's 2026 survey, the question that best predicts how a tech worker feels is: what has AI done to your sense of who you are?

About half feel amplified. They can do more, and better, and they're having fun. The other side feels destabilized or diminished, less sure what's really theirs anymore. That gap is the single strongest predictor of career optimism in the whole dataset, roughly three times larger than the famous "founders are happiest" effect. So the person sitting next to your happiest engineer might be the one about to quit, and their title won't tell you which is which.

The fear isn't robots. It's the pace.

Here's the part that should change how you run your next planning meeting. Only 22% worry about losing their job to AI. The bigger fears are being asked to do more for the same pay (51%), getting stuck in an unsustainable pace (46%), and watching the quality of their work drop (41%).

Meanwhile 82% say AI makes them measurably more productive, and burnout jumped 11 points in a year to 55.7%. Both things are true at once. People are shipping more and wearing out faster. When someone tells you AI made them faster, that is not the same as telling you they're okay. One Director of Product put the resentment plainly: "Use AI or you will lose your job, and then people get fired anyway. I hate it."

Your designers are waving the newcomers away

More than half of tech workers would steer someone starting out away from their own role. The average recommendation score is an NPS of -39. People made peace with their own path, but they've lost faith the on-ramp still works for the person behind them.

Designers and researchers sit at the bottom. They report the most AI anxiety, the most fear of job loss, the worst-rated managers, and the lowest willingness to recommend their field. That's a continuation, not a blip. If you run a design org, this is your problem before it's anyone else's. The people who mentor and hire juniors are the same people saying "don't come in," which is how a talent pipeline dries up two years before you notice.

The hiring lever is jammed too

Don't count on backfilling your way out of this. The market is a mess of crossed wires. Gergely Orosz found hiring managers drowning in AI-polished resumes while great senior people get ghosted. One firm gets about 1,000 inbound applications a day, and maybe two are relevant. Recent hires come almost entirely through networks.

So the plan of "if someone burns out, we'll just hire" is weaker than it looks. Inbound is noise. Cold applications from strong people go unread. That makes keeping the people you already have the cheaper move by a wide margin. Retention is now a hiring strategy.

The deep cut

Manager quality is still the strongest driver of burnout in the data, and that's the part you actually control. You can't fix AI. You can fix how your team feels the pace. That starts with giving people permission to push back. Kai Wong argues in UX Design that when "do more with less" hits, the path of least resistance is to keep your head down and ship whatever you're handed. That silence is where your good people quietly check out before they leave.

So build the muscle Petra Wille describes on Teresa Torres' podcast: saying no well, with evidence and clarity. If your team can say no to the fifth AI-accelerated ask this quarter without fear, they stay in the "amplified" half. If they can't, the extra speed just buys you a faster path to a resignation letter.

Three questions for your team

• Do you know which of your people feel amplified by AI and which feel destabilized? If you can't name them, ask, because their title won't tell you.
• Is your team's higher output turning into more shipped work at the same pay, or into slack they can actually use? One of those keeps people. The other burns them.
• When someone on your team pushes back on an AI-driven ask, does it cost them anything? If it does, you're training them to go silent, then go.