There's a debate doing the rounds at the moment, and it usually arrives in the form of a worried question. If AI can write the email, draft the proposal and answer the customer, are we slowly forgetting how to do those things ourselves? The motion on the table, more or less, is this: the spread of generative AI is making us worse thinkers.
I understand the worry. But I think it gets the history backwards. Every time we have built a new way to talk to machines, someone has warned that the previous skill was about to rot away. They were partly right and mostly wrong. The skill changed. The thinking moved up a level.
We've had this argument before
Look at how people have actually spoken to computers over the years. In the beginning you wrote machine code, raw instructions the processor could run. Then came Assembly, which was a little friendlier but still brutal. Then C and C++ let you stop worrying about exactly which register held which value. Then languages like C# and Python hid even more of the plumbing. Each step let you say more while typing less.
And at every step, the old guard grumbled. Real programmers write Assembly. If you use a language with garbage collection, you don't really understand memory. Graphical interfaces will make people lazy because they won't learn the command line. We have heard a version of this complaint for fifty years.
Here is the thing. Every one of those abstractions was, on paper, a loss of control. In practice, each one let far more people build far more useful things. The person who once spent a week hand managing memory could now spend that week solving the actual problem the software was meant to solve. Natural language, the way we now talk to AI, is just the next layer on that same staircase. You describe what you want. The machine handles more of the how.
What actually gets outsourced
The fear is that AI does our thinking for us. But look closely at what it tends to take off your plate, and it's rarely the thinking. It's the typing, the formatting, the remembering of which menu hides which setting, the third draft of a booking confirmation that sounds like the first two.
Running a service business in Australia, you know this better than anyone. The hard part of your week is not writing the words "thanks for your enquiry." The hard part is deciding which enquiries are worth chasing, what to charge, how to staff Saturday, whether that quiet Tuesday means you have a marketing problem or a pricing problem. That is judgement, and judgement is exactly the muscle that gets stronger when you stop spending your evenings on admin.
We see this in how people use Hixel. When the AI drafts a campaign or qualifies a lead or suggests next week's roster, it is not making the call for you. It hands you a starting point and asks you to decide. You still choose. You just choose faster, and you choose more often about things that actually matter. That is not less thinking. It's thinking spent better.
The real risk is laziness, not the tool
Now, I won't pretend there's no danger here. There is. The risk is not that AI thinks for you. The risk is that you let it, and you stop checking its work.
A calculator made arithmetic effortless, and yes, plenty of us lost the knack for long division. But the calculator didn't stop accountants from understanding money. It freed them to think about cash flow, tax and growth instead of grinding through sums. The people who got worse at thinking were the ones who treated the output as gospel and never asked whether the number made sense.
So the question to ask of generative AI is not "will this make me dumber." It's "am I still the one making the decision." If the AI drafts a price increase and you publish it without thinking about your margins or your regulars, that's on you, not the tool. If it suggests a roster and you sign off without noticing it put your newest hire on the busiest shift, the failure is human. The tool gives you a draft. You bring the judgement.
Aiming higher, not lower
There's a quiet snobbery in the idea that effort only counts when it's tedious. By that logic, the most virtuous business owner is the one hand coding their own website and stuffing envelopes at midnight. I don't buy it. Effort spent on plumbing is effort stolen from strategy.
What the better abstractions have always done is move the human up the ladder. From flipping switches to writing code. From writing code to describing outcomes. Each step asked less memorisation and more clarity about what you actually want. Knowing what you want, and being able to say it plainly, turns out to be a genuinely demanding skill. Anyone who has tried to brief a designer, or write a clear instruction for a new staff member, knows that.
So no, I don't think this house should believe that AI is degenerating our critical thinking. I think it's relocating it. The grunt work shrinks. The judgement grows. The businesses that struggle won't be the ones that used AI. They'll be the ones who switched off their brains and assumed the machine had it covered.
The machine doesn't have it covered. It never has. It just clears the desk so you can get to the work that was always yours to do.
