Every network engineer who reads this will probably back me up on this one. On the days we are not out in the field, our job looks almost identical to any other office job. We are stuck doing paperwork or staring at a monitor, which, if anything, works out in our favor. My coworkers and I have automated almost all of that paperwork with AI at this point, to the point where leaving early most days is now just normal.
But a thought hit me recently that I cannot shake. The more I lean on AI, the more it starts handling the exact things I always thought I was good at, and that brings on a weird, hard to explain kind of dread. I used to have to dig through the manual myself and summarize whatever I found. Now I just ask, and it pulls the data, compares it, and builds the table for me. If I tell it to format the report for a specific executive, it will even write it using the exact phrases that person likes to hear.
It still messes up plenty, and there are parts a human absolutely has to check. But what scares me more than where AI is right now is how fast it is moving. If it can already do this much, where does it land in three years? I genuinely wonder if the things I am good at right now will even matter by then. By that point, a lot of office jobs might just be gone, and even engineers like me could get replaced by humanoid robots that never need a paycheck.
Am I Just A Messenger Now
Like I said, on the days I stay off the field, I am basically glued to a monitor like everyone else in the office. And lately, every time I sit down to do paperwork, this hollow feeling creeps in. Whenever a question comes down from above, I ask the AI to dig up the data and organize it, then I take that and hand it over as my report. At some point it hit me. If this is all I am doing, how is it any different from the director just asking the AI directly?
 |
| The exact moment it clicks that you are the messenger, not the expert |
Strip it down to the simplest version, and here is the loop. The director asks me something, I ask the AI the same thing, I clean up the answer a little, then I hand it back. It leaves me wondering whether I am the messenger standing between the AI and the director, or just the guy typing on the AI's behalf. That thought genuinely rattled me. I imagine anyone stuck doing pure paperwork all day would feel this even harder than I do.
All of this kept eating at me and made me deeply anxious at first. It felt like my own value was shrinking a little more every single day. That is what pushed me to get serious about mastering ChatGPT and picking up automation tools like n8n. I even went down the same rabbit hole I described in How AI Natives Actually Use AI to Its Limits, trying to squeeze every last feature out of every tool I could find. I was convinced the only way to survive was becoming someone who could handle AI better than everyone else around me.
The Meta Is Shifting
But somewhere along the way, that whole mindset shifted for me. After running a bunch of AI experiments at work and weaving it into different parts of my job, I realized that was not the real answer. Sure, using AI well matters. But is getting really good at one specific tool an edge that lasts long term? Every time I finally learn a tool, an update rolls out and changes half of it. More than once, a feature I struggled to master turned into a default setting a few months later anyway. It is the same conclusion I already reached in Why Learning AI Tools Is a Waste of Your Time, that chasing tool mastery is a treadmill with no real finish line. At the end of the day, AI is just a tool, and tools tend to get designed so anyone can pick them up without much effort.
 |
| Every task looks the same until you break it into pieces |
Because of that, I think what matters far more than how well you operate the tool is knowing exactly what to ask it for in the first place. Writing a report is not really one task. It breaks down into several smaller ones, pulling the data, organizing it, structuring the outline, and reviewing the final draft. What really matters is splitting those pieces correctly, handing off whatever AI does better and holding onto whatever still needs a human.
The True Talent Of The AI Era
So here is where I have landed lately. Being good at AI has nothing to do with writing clever prompts. It comes down to knowing your own job cold enough to judge exactly what to hand off and what you still have to carry yourself. That, to me, is what real talent looks like in this era.
At the end of the day, planning itself has become a massive deal. That is why I keep trying, one task at a time, to hand more of my work off to AI. Instead of building some grand automation system from scratch, I am starting small, cutting down the repetitive stuff that eats my time, things like meeting notes, data cleanup, spreadsheet analysis, and first drafts of reports.
But what happens after that matters even more. Say AI cuts your report writing time from three hours down to thirty minutes. You do not spend that leftover two and a half hours cranking out another report. You spend it thinking, planning, doing a little more of whatever only a human can actually do. I think that is what eventually becomes the skill that counts. The ideas only a human comes up with, the human touch in the final product, making a report actually read like a person wrote it. My guess is these are the things that end up mattering most.
 |
| I don't have the answer, just the next task |
Of course, I have no idea if this is the right call. My job could vanish entirely because of AI, and nobody can really predict what our jobs will even look like ten years from now. But being good at paperwork the old fashioned way is clearly not going to be enough to stay competitive anymore.
So instead of sitting around dreading the day AI takes my job, I am going to be the one who hands off whatever it does better, one task at a time. Whatever time that frees up, I am putting toward planning, thinking, and judging what actually needs a human, then doing more of exactly that. Down the line, I plan to write about the specific tasks I have handed off, what genuinely worked, and what completely fell apart.
For deeper reads on tech, business, and the shifting economy, follow The Techtonic for weekly breakdowns.
Latest Insight: Why Corporate AI Training Doesn't Actually Work
Trending Insight: Stop Learning and Just Start Copying in the AI Age
Comment