AI Gave You Your Time Back. Take It.
by Lisa Peyton, full-time faculty at the University of Oregon School of Journalism and Communication, AI marketing practitioner, and Forbes contributor
I’m writing this from my sofa.
There’s a spider outside my window who has spent the better part of an hour meticulously weaving her web, my dog is stretched out next to me, and the weather is doing that thing Oregon does in August that makes you forgive it for February. I’m nowhere near my desk. I’m talking this post through out loud with Claude, the same way I’d talk it through with a colleague, and it’s going to be some of the best thinking I do all week.
Tomorrow, Friday, August 21st, I’m turning it all off.
The newsletter is taking a four-week break. The blog goes quiet. The teaching, the projects, the content machine, all of it goes on pause. I’ll keep a few small mandatory things running, but the bulk of my work is stopping, and I wanted you to hear it from me so you’re not wondering where I went.
Picking that date was hard…(but the relief I felt once I had made the decision was the only verification I needed that it was the right one.)
Hard because I know exactly what happens while I’m gone. New models drop. New capabilities land. The whole ground shifts a little, and I come back to a version of this industry that’s a few degrees different from the one I left. I open LinkedIn on a random Tuesday and there are forty posts in my feed, each one carrying a quiet little instruction: you should be testing this, you should be building that, you should have an opinion about this by Thursday. Multiply that by every platform, every newsletter, every release note. Unless I’m deeply intentional about what I let in, I get overwhelmed fast. And the overwhelm doesn’t make me work harder. It makes me feel paralyzed.
If that sounds familiar, you’re in good company, and I mean that literally. The research is starting to catch up to what a lot of us have been feeling. In Marketing Week’s 2026 Career & Salary Survey of 2,350 marketers, 65.3% said they’d felt overwhelmed in the past twelve months, and more than half reported emotional exhaustion.
Here’s the part I’m genuinely excited about. This is a young area of research, but it’s not an empty one. There are evidence-backed things we can do, starting today, to take care of ourselves while we work at this pace. I found nine of them, and I’ve laid them all out below.
I want to be a good example of this rather than a cautionary tale, so let me share my AI usage routine.
I’m up at five most mornings. Dog walk, then desk, and I’m there for a few hours before most people are awake. I cap my heavy AI work at two to three hours, because past that I’m wrung out in a way that’s hard to describe to anyone who hasn’t done it. The single best change I’ve made in the last year is that I now take Claude with me on walks. Dog on the leash, phone in my pocket, talking through strategy while my body moves. It’s better thinking than I do sitting down.
One warning on that, because I don’t want to hand you a bad idea dressed up as a good one. Walking-and-talking with your AI is still WORK. It is not a break. It’s the same cognitive load in a healthier posture, and if you count it as time off, you’ll wake up one day having gone eleven weeks without an actual day off. Ask me how I know.
Which brings me to the thing I’ve been saying for years now, and believe more every month: we ABSOLUTELY cannot keep pace with our AI companions. We were never built to. They don’t get tired, they don’t get overwhelmed, and they don’t need a spider and a sofa to think clearly. We do. So let them run, and let yourself be a human with a human body and a human nervous system.
If you’re using AI well (and I KNOW you are!), you’re producing five, six, seven times what you were producing five years ago. That’s worth celebrating. So take the day. Take the week. Take the month if you can swing it. And if you can’t take any of those yet, take the nine recommendations below and start there.
Sort the Work Before You Prompt
The study that named “AI brain fry” surveyed 1,488 workers and found the fatigue runs in one direction. Handing AI your repetitive work lowered burnout. Babysitting multiple AI systems raised it. For marketers, 26% reported brain fry, the highest of any profession studied.
So sort before you prompt. Three lanes: delegate the repetitive stuff (draft variations, formatting, summarizing material you supply), co-create the judgment work (concepts, positioning, editing), and keep the high-stakes work fully yours (claims, compliance, anything a lawyer would read). The delegate lane is where AI gives you energy back.
Write Your Routing Rules Down
You make the delegate-or-not call dozens of times a day, and experiments published in Information Systems Research found people make it badly on the fly. We’re unreliable judges of our own relative strengths. Human-AI teams beat the AI alone only when the division of labor was structured.
The fix takes ten minutes. Write your three lanes down, once, and stop re-deciding. A sticky note beats forty gut calls a day.
Three Tools Is the Ceiling
The brain fry researchers found productivity peaks at three simultaneous AI tools, then drops. Attention science explains why: Sophie Leroy’s experiments show every switch away from unfinished work leaves residue that degrades whatever you do next.
Pick your stack, keep it small, and batch your sessions. Draft in one block. Review in another. Retire the tools that overlap. Orchestrating multiple models is a superpower, and it has a hard limit per sitting.
Cap Your Sessions Before They Cap You
This one I’ve lived. I cap my heavy AI work at two to three hours because past that point, my thinking gets noisy and my body starts filing complaints. The research agrees that intense AI oversight is unusually depleting, and I’ve written a whole playbook on why the two-hour work day is your new power move. The short version: in two focused hours I can produce what used to take a week, and pushing past that produces worse work, not more of it.
Match the Break to the Work
A meta-analysis of 22 experiments found micro-breaks measurably restore energy and reduce fatigue. It also found the catch: after cognitively heavy work, ten minutes isn’t enough.
So scale it. Quick breaks between light tasks, real recovery after a deep review session. Step outside. Watch a spider build a web. And remember my warning from up top: talking to your AI on a walk is wonderful, and it’s still work.
Log Off Like You Mean It
The largest recovery study ever assembled, 316 samples and 99,329 participants, found that mentally disconnecting from work after hours reduces exhaustion and negative emotion. Not thinking about work is a skill, and it pays.
For us, that means no dashboard peeking, no “one quick prompt” at 9 pm, no monitoring the feeds for the model drop. The industry will shift while you’re gone. It shifts while you’re watching, too. Only one of those versions of you comes back rested.
Learning AI Is the Self-Care
My favorite finding in the whole research base. That same giant recovery study found “mastery experiences,” learning something new and rising to a positive challenge, improve both your wellbeing AND your performance. And a study of 416 professionals found that confidence in your own ability to learn AI weakens the link between AI adoption and stress.
Put those together: every hour you spend building real AI skill is doing double duty. It’s making you better at the work and buffering you against the pressure of it. You’re not choosing between excellence and taking care of yourself. Done right, they’re the same practice.
Decide How Hard You’ll Check Before You Start
Verifying AI output is real cognitive labor, and a systematic review of 35 studies found that checking gets worse exactly when workload and time pressure are high. Checking everything exhausts you. Checking nothing is how errors ship.
The answer is tiers, set before the work starts. Spot-check the low-stakes copy variants. Fully verify anything factual, regulated, or reputation-critical. One decision up front replaces a hundred anxious ones mid-task.
Ask What’s Being Subtracted
The strongest intervention evidence on record is about workload, not willpower. A 2023 meta-analysis of 13 controlled studies found that redesigning workload cut exhaustion more than any other approach tested.
You control this more than you think, because you control the question. When AI adds prompting, verifying, and coordinating to your plate, ask what’s coming off it. Ask it of your team, your boss, and most importantly yourself. Every capability you gain is a chance to subtract something, and subtraction is the highest-evidence move on this whole list.
Start With One
Pick a single practice this week. Write the three lanes on a sticky note. Cap tomorrow’s session. Take the real break. Then notice what changes.
And when you want to talk this through with an actual human, that’s what our community is for. Join my live monthly Advanced AI meetup, where you get to talk to a REAL person (me!) about what’s working, what’s challenging, and what to test next. While I’m away, the AI marketing resources hub and the back catalog are stocked and waiting.
See you in four weeks, my loves. Rested.
Made with my team of AI superheroes and my own skills, including AI-generated imagery. Every opinion, edit, and fact-check is mine. I am an AI practitioner, professor, and pioneer helping marketers put AI to work with purpose. Find more resources at lisapeyton.com/ai-marketing-resources or connect with me on LinkedIn.
