[AI SPRINT] AI Is Working. So Why Are Your Best People Burning Out?

I'm writing this from vacation. Which is fitting, given the topic.

Even here, I caught myself doing it. A quick AI session to research something surfaced three related questions. Which led to a framework. Which led to drafting a client deliverable I wasn't supposed to touch until next week. An hour gone before I noticed.

No one asked me to do that. I did it because it felt like momentum. Because the AI made it easy, interesting, and if I'm honest, genuinely fun.

That's the part nobody talks about. I’ll tell you about it, and the problem it causes, but a quick thing first:

I'm looking for 5 companies with 50+ employees to help our AI research. I've found a way to create your AI roadmap in as little as 10 minutes. Share your feedback and I'll give you that plus our full AI strategy assessment and debrief, at no cost. We normally charge $10,000+ for this. Interested? Just reply.

Now, onto the problem.

The Superpower Problem

AI-forward professionals are operating in something we've never seen before. You have an always-available AI team that gets meaningfully more capable almost every week. Tasks that required waiting on a colleague, hiring a specialist, or simply going without now get done in minutes. It feels like a superpower because it is one. And that feeling is exactly what makes the trap invisible.

What used to be a quick Google search now turns into something else entirely. You ask a question. AI surfaces five angles you hadn't considered. You pull the thread. Soon you have a research summary, a strategic plan, and the first draft of an implementation approach. Forty-five minutes later you look up and wonder where the morning went, not because someone demanded it, but because you felt engaged the whole time.

AI-forward people are now effectively always on, thinking, building, and exploring with no clear boundary around when work starts or stops. The work doesn't feel like burden. It feels like progress. That's what makes it so hard to step back from.

What the Research Is Starting to Show

In a Harvard Business Review piece published recently, researchers from UC Berkeley's Haas School of Business shared findings from an eight-month study following 200 employees at a U.S. tech company. Their conclusion complicates the story most AI vendors are selling: AI tools did not reduce work. They consistently intensified it.

Workers moved faster, took on a broader range of tasks, and extended their working hours, often without being asked or even noticing. If you work with developers, product managers, or designers who have gone deep on AI, you've likely already seen this. The PM who now writes her own code. The analyst who automated his reporting and immediately started fielding more complex executive questions. The designer who builds her own dashboards on top of her existing workload.

Each looks like progress on its own. Together, they point to something else.

The researchers identified a self-reinforcing cycle: AI accelerates tasks, which raises expectations of speed; higher expectations increase dependence on AI; dependence widens the range of tasks; widening the range increases the density of work. Repeat.

This is showing up first in tech-forward organizations that adopted AI early and went deep. But the pattern follows adoption. Whatever industry you're in, this is coming.

The harder problem is that most of this intensification is chosen, not imposed. Employees aren't being ordered to do more, they're drawn in by the capability of the tools. That makes it harder to detect and harder to manage. You won't see it on a status report. You'll see it twelve months from now when your best people start burning out or walking out.

The Choice Most Leaders Are Missing

When AI removes the low-value work, something fills the space. It always does. The question is what.

Most professionals default to more high-value work of the same kind they already do. More reports. More client calls. More transactions. More workstreams. It feels like productivity. And it is, right up until the cognitive load becomes unsustainable.

But there's a second option most people never stop to consider: using that freed capacity to deliver value you couldn't before.

The fork shows up everywhere, once you know to look for it.

A marketing team that used to spend most of their week on content production can now publish in a fraction of the time. Default move: more content. Strategic move: take ownership of demand generation strategy that previously lived with an outside agency.

A customer success manager who automated her renewal reporting suddenly has real availability. Default move: manage more accounts. Strategic move: shift toward proactive advisory work that deepens relationships and justifies higher contract value.

An HR team that automated job postings, screening, and onboarding can hire faster than ever. Default move: fill more roles. Strategic move: build the talent development function that never existed because there was never bandwidth for it.

And looking ahead, a real estate agent whose AI now handles paperwork, listings, and scheduling could ask a different question entirely: what might I now offer that I previously had to refer out? Basic investment analysis. Relocation consulting. Property management guidance. Services that were always adjacent but never reachable because of bandwidth.

That's the fork. More of the same at higher intensity, or something genuinely new. Most people never realize they're standing at it.

How to Take the Strategic Path

Here's the honest challenge: you probably can't measure what AI has freed up. The time doesn't sit idle, it gets absorbed instantly, invisibly, into more work. By the time you think to look for it, it's already gone.

So the actions that matter aren't about measuring what was gained. They're about interrupting the default path before it runs on autopilot. Here are three tips to avoid that:

1. Have the expansion conversation explicitly

Ask your team (or yourself) one question: what could we now offer or deliver that we couldn't a year ago? This conversation won't happen on its own. Put it on the calendar. Make it deliberate.

2. Watch for scope creep as a leadership signal

When people quietly absorb adjacent work, it looks like initiative. Often it is. But it's also how burnout builds invisibly. Check in regularly on what your team is actually working on versus what their role covers. The gap is where the risk lives.

3. Set intent before you open the tool

Before starting an AI session, name what you actually need: are you trying to fully solve something, or just inform a decision, guide someone, or scope the real work? Those require very different levels of effort. Without that clarity, the tool decides for you, and it will always pull you deeper.

4. Measure new value, not just volume

Introduce at least one metric that rewards something new delivered, a new service, a new capability, a new outcome. That's the signal that expansion is the goal, not just acceleration.

What This Means for You

The superpower feeling is real. The always-on thought partner, the daily new capabilities, the ability to move without waiting on anyone, that's not hype. It's a genuine shift in what's possible.

The trap isn't that AI makes work harder. It's that AI makes work so engaging and so frictionless that the boundary between working and not working quietly disappears, and most people don't notice until the cost shows up.

AI gives you capacity. What fills it is your strategy.

Hit reply and tell me: when you think about what AI has freed up on your team, or for yourself, what's actually filling that time right now? More of the same, or something genuinely new?

Trent Gillespie is CEO of Stellis AI and a keynote speaker helping business leaders understand and operationalize AI in their companies. He spent almost nine years leading global innovation efforts at Amazon before leaving to help other companies build the capabilities they need to compete. Book Trent to speak to your group or book a call to discuss using AI within your business.

Did someone forward this newsletter to you? If you're not already signed up, you can subscribe to AI SPRINT™ for free here.