When people say AI will destroy "laptop jobs", I think the question is framed incorrectly.
The laptop is just an interface.
What may disappear is a huge layer of work where the human was effectively the interface between different pieces of software.
Open Excel. Copy data from one system to another. Compare tables. Prepare a report. Move numbers into a presentation. Write a standard email. Find a document. Fill in a form.
Typing fast was once a profession.
Then typists largely disappeared.
But we did not stop writing.
Something similar may happen with Excel, CRM, SAP, browsers and dozens of office applications.
Excel does not need to disappear.
The human simply may no longer need to press every button personally.
Today:
human -> applications -> data -> result
Increasingly tomorrow:
human -> goal, constraints and responsibility -> AI -> applications, data and tools -> result
So the more interesting question is not:
"Will laptop jobs survive?"
It is:
"How much more will one person be able to do through a laptop?"
Today someone may prepare a report for a department.
Tomorrow the same person may ask:
"Why is our margin falling? Check purchasing, logistics, working hours, seasonality and planning errors. Show me three possible responses, their costs and their risks."
That is still a person sitting at a computer.
But it is not the same job.
The same applies outside the office.
On a construction site, AI can calculate materials, inspect estimates, compare photographs with plans, follow deliveries and flag deviations.
But somebody still has to understand why the wall is there, whether drilling in that place is safe, whether the work should be redone, who takes responsibility and which risk is acceptable.
The machine becomes more capable.
The useful human task moves higher.
So I would not say:
"AI will take laptop jobs."
I would put it differently:
AI will take a large part of the work humans currently perform inside interfaces.
And the laptop, phone, glasses or local AI system may become a much more powerful extension of the person using it.
A few years from now, the important question may no longer be:
"Which software can you use?"
It may be:
"What part of reality can you organize through it?"