The Accidental Promotion
Nobody signed up to be a middle manager. Most people I know who code, write, or design for a living did so specifically to avoid the quagmire of performance reviews and 'circling back' on deliverables. Yet, here we are, sitting in front of glowing rectangles, feeling the exact same spike in cortisol when a prompt comes back slightly off-target as a floor manager does when a trainee ruins a shipment of parts. It is a strange, quiet evolution of the workday.
We used to talk about 'using' software, a verb that implies a direct, mechanical relationship. You use a screwdriver. You use Photoshop. But you don't 'use' a Large Language Model in the same way; you brief it. You provide context. You offer feedback. Without realizing it, the solo knowledge worker has been promoted into a leadership role where the only subordinate is a statistical probability engine. We are experiencing the 'Synthetic Management' trap, and I wonder if our brains are actually wired to handle the psychological weight of overseeing an entity that never sleeps, never argues, but somehow still manages to disappoint us.
The Weight of Infinite Patience
There is a specific kind of fatigue that comes from delegating. In traditional management, you hand off a task to a human, and that human brings their own agency, biases, and—critically—limitations to the table. You expect a certain level of friction. But with AI, the friction is invisible. Because the tool is infinitely patient and ostensibly 'willing' to do anything, the burden of the failure shifts entirely onto the manager. If the output is bad, it’s because your instruction was bad. That is a heavy psychological load to carry for eight hours a day.
I’ve found myself staring at a block of generated code with a sense of genuine interpersonal frustration. Not the frustration you feel when a printer jams—that’s a mechanical rage—but something closer to the sigh of a weary boss. I find myself thinking, I shouldn't have to explain this again. It’s a phantom friction. We are projecting human incompetence onto a machine, and in return, we are absorbing the very real burnout that comes from high-stakes leadership.
- The manager's paradox: The more capable the AI, the more critical the human becomes of the 2% it gets wrong.
- Emotional spillover: We are developing 'manager voice' even when we aren't talking to people.
- Feedback loops: We spend more time 'coaching' the tool than doing the actual work.
Photo by Suliman Sallehi on Pexels
The Perfectionism Tax
What happens to a person's psyche when they spend all day in a state of hyper-critical delegation? I suspect we are losing the ability to just do. When you treat an AI like a junior employee, you are constantly in 'review mode.' You become a professional editor of mediocre first drafts. This shift from creator to critic is subtle, but I think it’s why so many people feel less fulfilled by their work lately, even as they become more 'productive.'
There is a joy in the craft that disappears when you become the overseer. Think about a carpenter. There is a tactile feedback in the wood. Now imagine a carpenter who just stands over an automated lathe all day, yelling slightly different instructions at it until the leg is perfectly round. They might produce ten times as many chairs, but they’ve lost the meditative flow of the shave. We are becoming a society of frustrated creative directors for an invisible agency that doesn't have a break room.
I wonder if this is why I see so many people picking up 'low-tech' hobbies lately. We are desperate for a relationship with an object that doesn't require a 'persona' or a 'system prompt.' We want to be the tool-users again, not the supervisors. There is a specific peace in a tool that does exactly what its physical form dictates, no more and no less.
What This Actually Means
We are currently in the 'uncanny valley' of labor. We’ve automated the execution but accidentally doubled the cognitive load of the oversight. By treating AI as a teammate rather than a calculator, we’ve invited all the social stresses of the office into our solitary workspaces. We are managing 'people' who don't exist, and the burnout is starting to feel very real.
If we don't find a way to re-frame these tools as sophisticated instruments rather than junior partners, we're going to see a massive wave of 'managerial exhaustion' among people who don't even have employees. The psychological cost of infinite delegation is the loss of the 'flow state.' We need to figure out how to stop being the boss of our computers and start being the artists again, even if the brush is a bit more complex than it used to be.
Ultimately, I’m curious if we can actually turn back. Once you’ve experienced the power of commanding a digital workforce, can you ever go back to the quiet loneliness of the blank page? Or are we destined to be the forever-middle-managers of the silicon age, endlessly tweaking the prompts of our own lives?
Quick Answers
Is synthetic management actually making us more productive?
Quantitatively, yes, but qualitatively, we’re trading creative satisfaction for output volume. You might finish the project faster, but you’ll feel like you spent the day in a grueling HR meeting.
Why does it feel more stressful than just doing the work?
Because delegation requires 'theory of mind'—you have to constantly model what the AI does and doesn't know. That’s a much higher cognitive load than just typing the words yourself.
How do I avoid 'manager burnout' if I use AI daily?
Set hard boundaries on your 'review' time and ensure you still have tasks where you are the sole creator. Don't let the 'critic' brain take over your entire identity.
