Probably not in the way most people are picturing. That is not optimism, it is a pattern: every major technology of the last century was expected to erase work, and the prediction failed in the same specific way each time. It was rarely wrong about what the technology could do. It was wrong about what people did next.
That distinction matters, because it is the difference between preparing for change and waiting for something that never arrives.
The post office was supposed to die
When email became normal, the conclusion looked obvious. Letters were the postal service's core business, and email was free, instant and effortless. Within a decade, people said, the post office would be a museum piece.
Half of that came true. Letter volumes fell across most of the world and have never recovered. Almost nobody posts a cheque or a birthday invitation any more.
But something else was happening on the same wire. The internet did not only replace letters. It created a way to buy goods from a warehouse four hundred kilometres away, and to order dinner from a restaurant that had never taken a delivery order in its life. Every one of those transactions ends the same way: something physical has to arrive at somebody's door.
The letter declined and the parcel expanded. Postal and courier work did not disappear. It changed shape, and in a great many countries it grew. The prediction correctly identified what was being destroyed and completely missed what was being built next to it.
The drone compressed a crew rather than replacing it
There is a cleaner example from film-making.
An aerial scene used to be a serious operation. It required a helicopter and a qualified pilot, a camera operator working from a stabilised mount, safety personnel, insurance, permits, a location that could accommodate an aircraft, and a budget only a real production could carry.
The drone reduced most of that to one person with a case in the back of a car. Not all of it, and it is worth being precise here: helicopters are still used for shots drones cannot achieve, and larger productions often run a two-person drone team with a pilot and a dedicated camera operator. The change was a compression of the crew rather than its elimination.
Even so, look at what followed. Some work genuinely disappeared. A pilot who made a living from aerial cinematography lost a client base, and that was a real loss to a real person.
Two other things happened at the same time. The roles merged rather than vanished, since a drone operator is flying, framing and shooting at once. More importantly, the cost of an aerial shot fell so far that people who could never have considered one began buying them: wedding videographers, estate agents, restaurants, farmers surveying their own land, small documentary makers.
The work did not shrink. It moved within reach of far more customers and multiplied. There are vastly more aerial shots filmed today than in the helicopter era, and vastly more people earning a living from them.
The pattern underneath
This recurs often enough to be worth naming.
The spreadsheet was expected to end accountancy. It did remove an enormous amount of manual ledger work, and bookkeeping clerk roles declined over the following decades. What grew alongside that decline was analytical work: accountants, auditors and financial analysts became more numerous, because once a calculation becomes almost free, organisations want far more calculations than before.
The cash machine is the example most often cited, and it deserves more care than it usually gets. Research by the economist James Bessen found that in the United States, teller employment continued to grow for roughly three decades after ATMs arrived, because machines made branches cheaper to operate and banks responded by opening more of them, while the teller's role shifted from counting notes toward sales and problem-solving. That is a genuine finding, but it describes a period rather than a permanent law. Teller numbers have been falling since around 2010, as online and mobile banking removed the reason to visit a branch at all.
That qualification belongs in any honest version of the pattern. Automation of a task does not reliably destroy a job, but neither does it protect one forever. What it does consistently is change what the job consists of, and change how many of them the market wants.
The mechanism is the same in each case. A task becomes automated. The cost of the output collapses. Demand for that output rises far enough that total human involvement holds steady or grows, while looking quite different from before.
Technology has been considerably better at removing tasks than at removing jobs. Most jobs are a bundle of a dozen or more tasks, and machines have historically taken three or four, leaving people to do more of what remains.
What is actually happening right now
The historical pattern is useful, but readers reasonably want to know what is visible today rather than in 1985.
The clearest early effects are on tasks that are repetitive, text-based and produced in volume: first-draft copy, routine translation, basic image and layout work, standard customer support replies, and the boilerplate portions of code. Freelancers at the entry level of those markets have felt it first, because the work most exposed is the work that was already commoditised and priced low.
What has not happened, so far, is a broad collapse in demand for the underlying professions. The pattern emerging looks familiar: the same people producing more, clients expecting faster turnaround, and the value moving toward whoever decides what should be made, judges whether it is any good, and takes responsibility for the result.
It is genuinely too early to be confident about where this settles. Anyone claiming precise numbers about jobs lost to AI is guessing, and the honest position is that the evidence is early, uneven across industries, and still arriving.
What is different this time
It would be an equal mistake to insist that nothing has changed.
Two features of this wave are genuinely new. The first is breadth. Earlier technologies affected one industry at a time, giving the wider labour market years to absorb the change, while this one touches writing, design, software, support and analysis simultaneously. The second is speed. The drone took most of a decade to reshape aerial filming. AI tools have changed working practices in some fields within months.
There is also a cost that optimistic accounts tend to skip past. When that helicopter pilot lost his aerial work, "the industry created more jobs overall" was no comfort to him. Aggregate figures do not pay an individual's rent. New roles frequently appear in different places, demand different skills, and open at a different stage of a career, and the person displaced is rarely the person hired.
So the defensible position is neither "nothing will change" nor "everything will be fine". It is that the destruction is visible and immediate while the creation is diffuse and slow, which is precisely why every generation predicts the end of work, and why the prediction has not yet held.
What this means in our region
Most writing on this subject assumes a Western labour market, and the picture across the Middle East and North Africa is not identical.
Two differences stand out. Adoption tends to arrive later and then move quickly, which compresses the adjustment period once it starts. And the composition of employment is different, with a larger share of public sector, construction, logistics, hospitality and field-based work, most of which is far less exposed to text-and-screen automation than the office roles that dominate these discussions.
There is an opportunity in it as well. Language has long been a barrier to remote work for the region's graduates, and translation tools have lowered that barrier considerably. The same technology that threatens some local roles also makes a candidate in Baghdad or Amman legible to an employer hiring remotely from anywhere.
What it means for you
Three conclusions seem reasonably safe.
Learn the tool that threatens your task. The drone did not put camera operators out of work. The camera operator who learned to fly one became the person everybody hired. The advantage consistently goes to people already in the field who adopt early, because they understand the work the tool is meant to serve.
Separate the task from the judgement. Automation takes the repeatable half first. If most of your day consists of filling in a template, that half is exposed. If most of it is deciding what belongs in the template, deciding who to trust, or handling a person who is upset, you are on considerably firmer ground.
Expect the title to change more than the work disappears. Most of the roles that will exist in ten years do not have names yet, in the same way that "drone pilot" was not a career when helicopter crews were busiest.
What it comes down to
Every wave of automation has produced two groups: people who watched the tool arrive and hoped it would go away, and people who picked it up early and became the ones operating it.
The technology itself has never decided anyone's fate. It changes what is cheap, and then people decide what to build with the savings. That has held from the printing press to the spreadsheet, and there is no strong reason to believe it stopped being true this year.
