Still Aimed at Humans

I’ve already written the version of this that keeps me up at night. When I sit down and honestly map out where all of this leads over the long run, I don’t find a clearly good road, and I’ve said so. So consider this the other half of an honest person’s ledger. Not a change of heart, a change of timeframe. Because when I stop staring at the far horizon and look at the next few years instead, at what actually happens to jobs and to the people doing them, I find something I can genuinely be hopeful about. I’ve been meaning to write it down.

Not the movie version

Let me set the frame, because both of the popular stories are wrong. AGI is not around the corner. We are not all going to be jobless, collecting a universal income, five years from now. That version sells headlines and raises money, and I don’t believe it. But the opposite story, the one getting louder as the bubble starts to wobble, that this is all stupid and nobody should touch any of it, is just as wrong. The technology is real and it’s useful, and pretending otherwise is its own kind of denial. The truth sits in the unglamorous middle: a genuinely powerful tool, arriving fast, that changes what work is without ending it.

And here is a fact that should cool the panic a little. The strategy everyone fears, replacing the humans wholesale, is mostly failing on its own. The pilots built to swap people for software keep dying, partly for ethical reasons but mostly for boring business ones. Half the time the tool turns out to cost about as much as the person it was meant to replace, so which was better? Companies are quietly discovering that the mass-replacement dream doesn’t even pencil out. That failure is buying us time to do the smarter thing.

Everyone moves up the stack

The smarter thing starts with a simple reframing: the machine takes the drudgery, and the human moves up. I watch this land the moment I make it concrete. Take someone in a finance department. For years they’ve been so buried in closing the books and wrangling reports that they never had a spare hour to think. Free up that hour and the questions change. What reports could you run that you never had time to build? What insight is sitting in the data that nobody has looked at? What small tool would give the leadership team a view of the business they’ve never had? And bigger than that: finance people, once they’re not drowning, are often the ones who could help sales, or help manufacturing, if they could ever get out of their own silo. The busywork is the wall. Take the wall down and people don’t run out of things to do. They start doing the things that were always above their pay grade and never within their reach.

That’s the pattern in every department. The drudgery was never the job. It was the tax you paid to do the job. Stop paying the tax and what’s left is the part that needed a human all along: the judgment, the taste, the deciding what is even worth doing.

The shape of the new work

If people are moving up, up into what? This is where the org chart quietly gets redrawn. One way I’ve started describing it: treat any function in a company as a product, and a product needs a handful of distinct jobs done. Someone to generate the raw ideas and the rough prototypes. Someone to weed the weak ones out and push the promising ones forward. Someone to take an idea and actually productize it, and decide how you’d know if it worked. Someone to harden it and make sure it fits the people who will use it. Someone to grow it. Those roles exist whether you’re talking about a marketing campaign, a finance dashboard, or a whole new service line. What changes is that the tools now let far more people take part in that chain, so the work reorganizes around ideas and ownership instead of around who can grind out the manual steps. In a real sense everybody becomes a manager: of tools, of small projects, of their own little product, whether or not it says so on the badge.

None of that is a world with less work in it. It’s a world with different work in it, tilted toward the parts people are actually good at and away from the parts they only ever did because nobody had invented a better way.

Why the human doesn’t go away

The deepest reason I’m not worried about humans vanishing from the picture is almost too obvious to say out loud, which is exactly why people skip past it. Everything we build is still aimed at other humans. Every product, every service, every campaign, every interface, at the far end of it is a person, with a bad day and a budget and a life you have to actually understand in order to serve. The machines can generate and summarize and optimize all day, but the judgment about what is worth making, and the real feel for how a specific human being experiences the thing you made, stays stubbornly human. You can automate the production of an answer. You cannot automate caring whether it’s the right one for the person in front of you.

“It’s all still aimed at humans.”

That’s the part that doesn’t compress. The better idea, the truer read of someone’s daily reality, the taste to know that technically-correct and actually-good are two different things: those remain the human contribution, and if anything they get more valuable as the mechanical parts get cheap. Scarcity moves to wherever the machine is weakest, and the machine is weakest precisely where being a person matters most.

Write your own story

There’s a smaller, more personal reason I lean hopeful, and it’s the one I’d want a nervous person starting out to hear. Right now, for about the price of a couple of coffees a month, anyone with curiosity and a little nerve can pick up tools that would have needed a whole team a few years ago. If you’re young and ingenious, you can build things, test things, and bring them to your job on Monday. You can rethink how your corner of the company works and simply show people. You get, more or less for free, an education in the most consequential technology of the moment, and you can put it on your resume. That is not a threat to a curious person. It’s the best leverage a beginner has ever been handed. The ones who treat this as a chance to write their own story, rather than wait to be written out of someone else’s, are going to be fine, and a few of them are going to do something extraordinary.

Two honest caveats

I owe you two pieces of honesty, or this turns into exactly the kind of hype I can’t stand.

The first: not everyone wants this. There are people, plenty of them, who like clocking in, doing defined work, and clocking out with their mind free for the life that happens after. “Become the creative brain of your department” is not a gift to someone who never wanted their job to colonize their thoughts. A humane version of this future has to have room for them too, for the person who just wants good, steady, honest work that pays the bills and then leaves them alone. Hopefully we can make it good for most people. I mean that as a real aspiration, not a slogan.

The second: I haven’t taken back a word of the worried version. The long-run map still troubles me, and I still think minimizing harm may be the most honest goal we have. But here’s how I hold both at once. The far future is a set of scenarios I can influence and not control. The next few years are a set of choices I can actually make, with clients and with the people I work alongside, about whether we point these tools at subtracting humans or at freeing them. Brightness, in that sense, isn’t a prediction. It’s a decision about where to aim. And on the near ground, where the aiming actually happens, I choose the version where the drudgery goes to the machine and the human gets to do the part that was always the point.

It’s all still aimed at us. We should build it like we know that.