Your company raised the bar because of AI (now you're burning out)
"I'm expected to do more now."
That's a direct quote from a developer in a peer-reviewed study presented at ICSE 2026 in April. Not a complaint about AI tools being bad. A complaint about what happened after the company adopted them.
The company got AI. Then management raised the bar. Now the developer has AI-level output expectations with a human-level body.
What the study actually found
Researchers surveyed 442 software developers across 56 organizations and modeled the relationship between GenAI adoption, job demands, job resources, and burnout.
The finding: GenAI adoption is strongly associated with increased burnout - but not for the reasons you'd expect.
The mechanism is job demands. When a company rolls out AI tools, two things happen simultaneously. Developers get assigned more tasks because "you have AI now." And they spend extra time debugging AI errors - output that doesn't work, code that looks right but isn't, suggestions that need full rewrites. The tools create work while also being used to justify more work.
Job demands were the strongest predictor of burnout in the model (β=0.398, p<.001). That number explains nearly 40% of the burnout variance across all 442 developers.
The support that wasn't there
Only 30.2% of developers received formal training or workshops for the AI tools their companies deployed. Another 22.4% reported no meaningful organizational support at all.
So: new expectations, new tools, almost no investment in helping people actually use those tools well. That's the setup for burnout.
The study also found something important on the protective side. Job resources - specifically autonomy - were the strongest buffer against burnout (β=-0.360, p<.001). Developers who had control over how they worked, how they structured their day, when they took breaks, burned out significantly less than those who didn't.
Larger organizations gave developers fewer choices about their workflow. Developers in coding-focused roles under the most scrutiny had the highest organizational pressure scores.
Autonomy is the actual fix
The conventional advice after a burnout study like this is "take breaks" or "set boundaries." That's not wrong, but it misses the mechanism.
The reason breaks help is that they're one of the few forms of autonomy left in a system that's constantly raising the output bar. Stepping away from the screen for 5 minutes isn't just a rest - it's a signal that you control your schedule, not the backlog.
That's what Movedoro does. It forces scheduled movement breaks into your workday whether the sprint is on fire or not. You don't decide in the moment whether you deserve a break today. The break happens, you move, and you come back.
It's a small thing. But when the company has raised every other bar, having one part of your workday that's non-negotiable and yours matters more than it sounds.
That's pretty much it.
