You Too Could Be Replaced by Predictive Text

It’s 2026, so we get to be anxious about being replaced by a computer.

Words by Amelia Luke

‘Artificial Intelligence’ is the buzzword of the decade, forcibly installed in everything from your computer to your phone to your toaster. But, of course, artificial Intelligence isn’t a real thing—it may be artificial, but it’s not intelligent. What is referred to as AI is typically what is actually known as an LLM, a ‘Large Language Model’, which probabilistically puts one word in front of another. It can’t think and it doesn’t even know what it’s saying – it just knows what the most likely combination of words to output based on the words you input. It’s literally the same as the predictive text on your phone, on an enormous scale, powered by titanic amounts of water consumption and electricity.

Nevertheless, predictive text is coming for your job.

To believe proponents of so-called AI, it’s a technology that can replace anyone. No white-collar job is safe. No creative job is safe. In their ideal world, no job is safe (except, of course, middle managers and executives).

Who needs a human artist or graphic designer? We stole all your art, and now a machine can clumsily recreate it (no, your labour will not be compensated). Who needs a human writer? We fed all your books and articles into a machine, and now it can clumsily recreate them (no, your labour will not be compensated). Who needs your average white-collar office worker anymore, when an AI can do any and all menial office work – or can it?

There’s a big difference between wanting AI to replace a job and it actually being able to do that job well.

In some industries, the idea that you could truly be replaced by an AI is rather spurious. Amazon workers have spoken about how much of their jobs have become fixing problems caused by being forced to integrate the company’s internal AI into their workflow, lowering their own productivity. This is a persistent problem at many software companies, where AI produces bad code that human coders then have to spend time finding and fixing. That hasn’t stopped thousands of jobs being cut at tech firms the world over, of course.

In other industries, however, AI supplanting one’s job is more existential. Take, for example, voice acting, where a voice actor’s existing work can be fed into a machine to create an artificial facsimile instead of hiring the original actor – an act which is not currently illegal. Is that better than hiring a talented and trained professional? Well, it’s certainly cheaper.

In schools and universities, AI threatens to replace students – if you have ChatGPT do all your work, are you even really there? Despite being an easy way to skip doing an assignment, recent research shows that using LLMs in education is an act of ‘cognitive offloading’ that impairs one’s ability to learn and retain information. AI use literally makes you less intelligent, as it degrades your ability to retain information and problem-solve. This is particularly insidious in the Humanities – if you aren’t learning to critically think and analyse in your Arts degree, you might as well be setting your money on fire.

AI replacing humans can be philosophically existential too. If an AI commits mass murder, such as with America and Israel testing autonomous killing technology in their war against Iran, does that mean no one can be held responsible? Oops, the computer blew up a school. Mistakes happen.

With that all said: Are you being lied to?

Neal Woolrich, from the research and advisory firm Gartner, is quoted by The Guardian as sceptical that AI is really replacing jobs at all:

“I think there’s a lot of use of AI as cover for other things that are going on in the organisation. We did some economic modelling last year and found only 1% of job cuts were the result of AI productivity gains. […] Sometimes when organisations are going through headcount reductions, there were other financial pressures that were driving it. I suspect there’s something else going on.”

In The AI Con, Drs. Hanna and Bender observe the phrase ‘artificial intelligence’ as being “deployed when the people building or selling a particular set of technologies will profit from getting others to believe that their technology is similar to humans, able to do things that, in fact, intrinsically require human judgment, perception, or creativity”.

AI proponents want you to believe their technology is far more advanced than it actually is because, when they use it to do whatever they want, they get to wash their hands of responsibility.

“The AI decided this was the most efficient thing to do.”

“The AI is more efficient than you.”

“We can’t be held responsible (for the decision we wanted to make).” 

Is AI hype really just another way for a company to justify downsizing? Firing a bunch of people makes the next quarterly financial report look so much better, after all. And don’t worry too much, if this whole AI thing doesn’t work out, the company can just hire you (or someone like you) back later – for less pay, of course.

Is there anything you can do in the face of all this? Well, you can at least not buy what they’re selling – your labour.