How Might AI Disrupt Our Jobs? Here are My 10 Reflections
Since the 2022 launch of ChatGPT and its amazing ability to generate content, an increasingly urgent question is what AI will do to our jobs.
Will Artificial Intelligence take our jobs? Will it merely augment and change the way we do our jobs? How do we even begin to think about this confronting topic in a way that avoids catastrophising, and yet is clear-eyed enough to see the challenges that lay before us?
I’ve been thinking about this topic a lot over the past year (not least because I’ll be presenting on this topic at a public lecture for the Centre for Christian Living, Moore College). And this is one issue that could keep me awake at night if I let it. But as Christians, we’re not victims of circumstance: we know that our great God is working out all things according to his good and gracious plans (Eph 1:11), giving us hope in Christ that nothing – not even AI disruption – can take from us.
With that in mind, here are 10 of my thoughts on how AI might disrupt our jobs:
1) Technology has always disrupted jobs, but AI is likely to disrupt jobs more than any technology before it, and faster
Since at least the invention of the wheel, technology has been disrupting our labour.
Technology can make certain jobs easier and faster (I can only imagine how the wheel revolutionised life for our ancestors), thus making some jobs and tasks redundant. However many observers are saying that AI is likely to disrupt and revolutionise the job market in ways we haven’t experienced before.
As AI company founder and author Mustafa Sulyman argues in his recent book The Coming Wave,
‘AI’s rate of improvement is well beyond exponential, and there appears no ceiling in sight. Machines are rapidly imitating all kinds of human abilities, from vision to speech and language…new language models can read, synthesize, and generate eye-wateringly accurate and highly useful text. There are literally hundreds of [jobs] where this single skill alone is the core requirement, and yet there is so much more to come from AI.’[1]
2) A majority of AI today is driven at an accelerated pace because it can be built before we decide if it should be
ChatGPT was released into the world without anyone’s permission or request.
AI startups are being given billions of dollars in investment capital to solve problems and automate jobs without society’s approval or permission. Of course, technology has always been let loose on the world, especially since the Industrial Era (steam engines, automobiles, aeroplanes etc). But technology is now being built and iterated faster than humanity can keep up with it.
3) The pace of AI development is unlikely to slow down anytime soon
AI researcher and Wharton School Professor Ethan Mollick recently summarised the pace of AI:
To be clear, nobody can tell you the future of AI accurately, except that AI development seems to be happening much, much faster than even experts expected…The average estimated date for when AI could beat humans at every possible task shifted dramatically, moving from 2060 to 2047—a decrease of 13 years—in just the past year alone! (And the collective estimate was that there was a 10% chance that it would happen by 2027).
And the investment into AI is only likely to keep growing.
According to the Wall Street Journal, OpenAI CEO Sam Altman recently announced his desire to raise $7 trillion (yes, trillion) to ‘boost the world’s [computer] chip-building capacity [to]expand its ability to power AI’. (In case you’re unfamiliar, AI needs super advanced computer chips to power it). The plan is to ‘solve constraints to OpenAI’s growth, including the scarcity of pricey AI chips required to train large language models behind AI systems such as ChatGPT.’ (Altman has since tweeted that he wants to increase that $7 trillion to $8 trillion).
4) The people most at threat from Generative AI (e.g. ChatGPT) are white-collar professionals
While earlier rounds of automation mostly affected blue-collar workers (although many white-collar workers were also affected, e.g. travel agents), the new generative AI mostly threatens white-collar jobs, as this WSJ article points out:
‘Generative AI could soon upend a much bigger share of white-collar jobs, including middle and high-level managers, according to company consultants and executives. Unlike previous waves of automation technology, generative AI doesn't just speed up tasks or recognize data patterns. It has the power to create content and synthesize ideas — in essence, the kind of knowledge work millions of people now do behind computers.’
What does this look like in practice?
Several freelance copywriters were finding that work was drying up as clients turned to generative AI programs like ChatGPT to do copywriting work (although now it has bounced back as clients realise that ChatGPT can’t yet do as good a job as humans).
5) There are different schools of thought on how AI disruption will work
While nobody knows for sure exactly how AI will disrupt jobs, there seems to be three main schools of thought that I’ve come across:
a. AI will disrupt, but create new jobs (like other periods of technological change)
This was the view of economist John Maynard Keynes, who coined the term ‘technological unemployment.[2] In Keynes's view, this was a good thing, with technological advancement leading to increasing productivity, which in turn frees up time for further innovation and leisure.[3] We’ve seen much technological unemployment over the last few hundred years: lightbulb factories did great as candlemakers went bust; car dealers came of age as saddle makers shrunk to a small niche industry. And of course, the middle-class rose on the back of white-collar jobs such as accounting and management that came because of industrial automation.
b. AI will disrupt jobs, but will augment human capabilities
This view of ‘human in the loop’ AI is the most positive that I’ve come across: AI augments human workers, which disrupts but doesn’t replace human workers (although the augmentation will mean fewer people are required to do the same amount of work). Thus, AI does not replace copywriters so much as helped by AI to do more copywriting, faster.
c. AI will disrupt jobs, and eventually replace most human capabilities
This is the most pessimistic (or perhaps realistic?) view of where AI is headed in the long term. No, AI isn’t going to take your job tomorrow, but in a decade the workplace will look quite different. As Sulyman explains:
‘[AI] will only temporarily augment human intelligence. They make us smarter and more efficient for a time, and will unlock enormous amounts of economic growth, but they are fundamentally labor replacing. They will eventually do cognitive labor more efficiently and more cheaply than many people working [in white collar jobs].’[4]
6) The Big Question: If AI disrupts and replaces workers, will new jobs come fast enough in the short term?
We’re now living at a time when most people in the West have adjusted to the world of industrialisation, computers, and the internet, and all the disruption that it wrought.
But with AI disruption on our doorstep, will new jobs come fast enough to replace those lost to human augmentation, or lost to AI outright?
As author Jeremy Peckham argues, this is an uncertain question:
Although economies adjust to technological shocks, the transition period is measured in decades, not years, and the rising prosperity may not be shared by all.[5]
Time will tell.
7) The 'Best Case’ Scenario for some analysts is increased leisure time as machines do our jobs. But even this is problematic
While some AI researchers and commentators look forward to the possibility of AI taking our jobs, and paying for humanity to live a life of leisure (in the same way higher classes in the Greco-Roman world were freed up for leisure by a class of slaves), we’re not designed to live a life of non-stop leisure.
As God’s image bearers, we’re designed for work (Gen 1:26-31; 2:15), and this is a requirement and responsibility affirmed throughout Scripture (e.g. in New Testament passages such as 1 Tim 5:8 and Titus 3:14). While rest and leisure are important for humanity – as seen in the institution of the Sabbath for the God’s people in the Old Testament, it is not something we’re meant to do fulltime (contrary to our holiday and retirement obsessed Western culture).
Our humanity is diminished if we can’t do meaningful work – whether paid or unpaid (e.g., volunteering or looking after children at home).
8) Technology is not determinative: society can respond to how AI is implemented
University of NSW AI researcher Toby Walsh points out that while technology like AI can and does shape society, it does not determine the future of society. As a society, we do have (some) choice as to how to implement AI technology.
We can do this through our governments (through labour laws and regulations), such as NSW state high schools recently banning students from accessing mobile phones at school. When I was in Budapest recently, Hungarian Taxi drivers were able to convince their government to ban the ride-sharing app Uber from Hungary.
We can do it in the workplace, especially if you’re in management. Are you going to use AI to augment your people, freeing them up for other high-value work, or will you replace your people with AI?
And of course, we can do it personally: how will we as families and individuals use AI? To shortcut learning through ChatGPT? Or to help us learn and research, and even build family businesses?
9) How should Christians respond to these changes?
The key question Christians need to ask when it comes to all technology, including AI, is this:
How do we make the most of AI (for the good of our neighbour, and the glory of God), while mitigating the negative effects of AI?
Again, we can do this at several levels: at the societal level, we can inform ourselves of AI, and engage politically (e.g. contacting your local member). At the workplace level, we can test and experiment with AI use cases. (And if your workplace doesn’t yet have an AI policy (what’s your management doing?), then push for that as well). And be sure to test and experiment with AI personally, and at home. Sometimes you will realise that AI is useful in some settings; in other settings, AI might need to be restricted, or even banned.
Either way, as Christians, we can and should voice the importance and dignity of human work. As Jeremy Peckham argues:
We can be those who strive to find a balance between technology and humanity, erring on the side of caution and preserving rather than eliminating work, with the yet unknown and perhaps unintended consequences to civilisation of such an elimination.[6]
10) The jobs least likely to be automated are those requiring high levels of human contact, manual skill, and/or constantly changing environments
Finally, it’s worth noting that the jobs that are least likely to be automated (at least for the near to medium term future) are those where people demand human contact, require high levels of manual skill, and/or deal with constantly changing situations.
Think kindergarten teachers (and those working with children in general – I don’t think many parents will want their kids taught by machines); clergy; psychologists and GP’s (although some people might be happy with online chatbots); coaches of sports (e.g. Karate instructors), tradies (plumbers, electricians, window washers), and so on.
[1] Mustafa Sulyman and Michael Baskar, The Coming Wave – AI, Power and the 21st Century’s Greatest Dilemma (Great Britain, The Bodley Head: 2023), 179.
[2] Sulyman and Bhaskar, 178.
[3] Sulyman and Bhaskar, 178.
[4] Sulyman and Bhaskar, 178. Emphasis added.
[5] Jeremy Peckham, Masters or Slaves? AI and the future of humanity (London, IVP: 2021), 148.
[6] Peckham, Masters or Slaves?, 157.