AI optimists are living in Rainbow Land

Yes, AI is going to take your job. The only question now is when that happens

I don’t want to live in Rainbow Land and you can’t make me!
The Campaign, 2012

Without bold decisions, the prospect of greater poverty and inequality looms large, which would leave many individuals marginalized, stranded and surrounded by the machines and automated systems that have replaced them… It is not enough to react only when jobs disappear; we must oversee the transformation in advance.
POPE LEO XIV, MAGNIFICA HUMANITAS: On safeguarding the human person in the time of artificial intelligence, 25/5/26

“I mean, driverless cars are writ large, coming fast in the US, and AI technology and robotics are replacing people, and it’s going to be on a mass scale,” [former Australian treasurer and US ambassador Joe Hockey] told the National Press Club.

Hockey urged governments to slash spending to make space for supporting such an army of unemployed. But there’s literally no way any government could cut so much out of the federal budget without driving the country into recession and precipitating riots on the streets.

This new technology is being dropped on to a world that was already splintering because of community-wide economic tension… The rise among young people who believe democracy is failing, as they engage with a terrible mix of fascism and communism, shows it’s not just older voters angry and bewildered by what’s going on around them…

Artificial intelligence offers extraordinary economic opportunities that are already evident in this country. But if that future is really as bleak as Pope Leo fears, then we’re about to take a step into the darkness.
Shane Wright, SMH, 3/6/26

AI is going to take your job, dear reader. Not necessarily in the next week, month or even year, but probably a lot sooner than you currently imagine.

Even the AI boomers no longer argue much against this proposition. They just insist – based on past technological disruptions – that AI will ultimately create as many jobs as it takes away.

Some of those making this argument may sincerely believe it. But as many prominent tech industry figures – Sam Altman, Dario Amodei, Bill Gates, Mo Gawdat, Geoffrey Hinton, Vinod Khosla, Kai-Fu Lee, Elon Musk, Mustafa Suleyman – regularly point out, AI is inevitably going to displace human labour.

As an erstwhile print journalist, I’m all too familiar with how humans typically react to a threat of a technological disruption that’s going to turn their life upside down. In my experience, they remain in denial about what’s clearly happening for an unfeasibly long period of time before cycling through the other destinations on the five-stages-of-grief journey – anger, bargaining, depression and acceptance.

That many journalists chose to keep living in Rainbow Land for years rather than facing the grim reality didn’t much impact anybody other than members of the Fourth Estate. Just like previous technological disruptions tended not to impact anybody except those directly affected – telephone switchboard operators, photo lab technicians, video store owners and so on.

But AI is close to automating away cognitive labour at scale.

Microsoft’s AI head honcho, Mustafa Suleyman, expects AI to achieve “human-level performance on most, if not all, professional tasks” by mid-2027.

Automating physical labour is going to take a bit longer, but not that much longer. Much of the work once done by humans in factories and warehouses has already been automated away. Basic and still pricey humanoid robots are likely to be a common sight by the end of the decade. From around 2030 onwards, there will probably be increasingly sophisticated and affordable general-purpose robots.

In short, if you’re a white-collar worker, you’ll probably find yourself out of a job and unable to secure another one within the next couple of years. If you’re a blue-collar or pink-collar worker, you’ve probably got a little longer – maybe 4-8 years – before the robots come for your job.

That’s not a lot of time to profoundly re-engineer a society. As the Vicar of Christ recently observed, it would be wise to start preparing for – at the very least – significant labour market disruption before it arrives in force. Especially in societies “already splintering because of community-wide economic tension”.

Granted, serious people – politicians, captains of industry, union bosses and now religious leaders – are talking about the need to fundamentally re-engineer societies. But none of them yet appear to be taking serious action to prepare for what’s coming.

That’s partly because a critical mass of people prefer to continue living in Rainbow Land. In Rainbow Land, the jobs for humans either don’t go away or are replaced by more enjoyable and better-paid ones.

This is what the young people call cope.

The three varieties of cope
With a tip of my hat to both Jasmine Sun and David Shapiro (well worth a follow if you’re interested in AI), let me explore how flimsy the straws AI optimists clutch at are.

The augmentation theory
This argument is true in the short-term but, unfortunately, not so much over the longer term. Most white-collar workers are already using AI as a tool. As flagged, many will continue to be AI-augmented workers for another year or two.

But what happens when a human in the loop is no longer necessary? What if there was some form of AI – let’s call it an AI agent – that allowed employers to get rid of the expensive employee? As one of Australia’s more colourful tech gazillionaires – who is in the process of firing a third of his workforce – recently observed, “It doesn’t take much effort to convince people, in the end, that they’re stupid to be paying $100 for labour when you can pay $2 for the AI.”

2) Jevons Paradox
The Jevons Paradox posits that when a technology makes a resource cheaper or more efficient to use, people may end up using more of that resource, not less. The classic example is coal. When steam engines became more efficient in 19^th-century Britain, factories needed less coal. You might expect that to have led to reduced demand for coal, but as coal-powered production became cheaper, more businesses invested in steam engines and demand for coal rose.

AI optimists argue a similar process will play out with AI. That is, it will make many services much cheaper, which will make them more accessible, which will increase demand, which will generate lots of new jobs for humans.

To give the AI optimists their due, the Jevons Paradox is a real phenomenon and there are plenty of examples of it playing out. But here we get back to AI being different from previous technologies.

Let’s say demand for software soars as it becomes less expensive. Surely, that should result in increased demand for human coders?

Well, only in an environment where (a) AI can’t write the code itself, (b) AI can’t help those who haven’t slogged through a computer science degree to vibe code and (c) AI doesn’t make senior coders vastly more productive than they were previously.

3) The humans-prefer-humans argument
The proposition here is that as more jobs are automated, workers will simply flow into industries where human-to-human interactions are currently considered important – therapy, tutoring, health and aged care, and personal services, for example.

It’s true that this dynamic has been playing out for many decades. Only 150 years ago, the majority of the population in Anglosphere societies were farmers or agricultural labourers and there were vanishingly few therapists, tutors, aged-care workers, nurses or personal-service providers. Nowadays, agriculture accounts for 1-3 per cent of the workforce and healthcare for over 10 per cent of it in first-world societies.

I suspect the ‘human premium’ will continue to exist for a while. For instance, low-to-mid-priced eateries will continue to automate by forcing patrons to order their meals via their phone or a self-order kiosk and having mini-Daleks transport meals to diners. When affordable, general-purpose robots arrive, they’ll automate the food preparation as well.

But I imagine the classier joints will continue to do things the human-labour-intensive way with waiters, sommeliers and perhaps a celebrity chef. Paying a human to do something a machine can do for a fraction of the price will almost certainly become a status marker. And whatever else happens in the coming years, I don’t imagine humans are going to become any less obsessed with status.

But if they are surviving on a near subsistence-level income from a UBI and/or gig work, most of the population simply won’t have the option of splashing out on human-provided services, no matter how status-enhancing it becomes to do so.

Plus, I’m not even sure people who can afford to employ humans will continue to want to do so. They might be willing to shell out on, say, a fancy dinner, but are they going to opt for the cashier over the self-checkout?

Let me quote the aforementioned Jasmine Sun:

I am currently a freelancer with limited access to costly human services. When my human therapist doubled her hourly rate last year, I decided to stop therapy and chat more with Claude instead…

The human premium is simply less significant than most assume. Plenty of people—including the wealthy—prefer Waymos to Ubers, telehealth to doctors’ visits, and Netflix to live theatre. Digital services often beat humans in convenience, consistency, and quality. And sure, artisanal ceramicists/bookbinders/basket-weavers today command high prices for their craft, but these are dwindling niches rather than mass-employment sectors.

Believe your lying eyes
Let me wrap up by posing a simple question – if AI is creating new jobs, where exactly are they?

The aforementioned David Shapiro explored this issue recently in a post titled Job cope is the opiate of the masses. After noting that Jensen Huang, David Sacks and Marc Andreessen keep insisting “AI isn’t causing mass layoffs!”, Shapiro writes:

First, AI is actually causing layoffs. Second, AI is causing entry level jobs to dry up. Third, the investment into AI is not creating that many jobs. This is the main point to ask: Where are the jobs, then?

Here’s the thing, you can use some simple rhetorical judo to say “Ah, well, the jobs just haven’t manifested yet!” Okay, but hundreds of billions of dollars are being poured into AI. In any other industry that would create jobs. So where are the jobs?

Indeed. Here are some other pertinent questions that Shapiro didn’t pose that I will.

Are you acquainted with any recent graduates who are crowing about the strong jobs market they are entering?

Do you know any business owners who are planning to grow rather than shrink the size of their workforce for AI-related reasons?

Have you heard any CEOs talking about how they’ll need to start staffing up soon, given the increased demand for human labour that AI is about to generate?

When was the last time you heard an open-border enthusiast arguing that Anglosphere societies will need to ramp up mass migration even further given the enormous labour shortages that AI will inevitably create?

Are electorates acting as might be expected if a golden age of full-employment, rapidly rising incomes and general abundance was just around the corner?

Why should we assume demand for human labour will rise when the explicit – albeit sometimes sotto voce – sales pitch of AI is that it will reduce the demand for human labour?

A time for bold decisions
Maybe AI optimists such as Andreessen, Huang and Sacks, will be proved right. Perhaps AI will create unexpected new industries, new roles and new forms of abundance.

But right now, the available evidence points to a bleaker future – fewer entry-level jobs, leaner workforces, ever less bargaining power for workers, and a widening divide between those who own the AI and those rendered obsolescent by it.