How safe is my job in the AI age?
On average, workers can expect that two-fifths (39%) of their existing skill sets will be transformed or become outdated over the 2025-2030 period. File photo
If you are one of thousands of workers currently worrying about your job in the age of artificial intelligence, then the utterances of some of its creators will only add to that anxiety.
Regarded as the godfather of AI, Geoffrey Hinton firmly believes that job displacement from AI is “more probable” than job replacement.
“If I worked in a call centre, I would be terrified," Hinton said. “It is already getting hard for university graduates to get jobs and part of that may be that people are using AI for the jobs they would have got.”
Asked “What would you tell your kids to do?” Hinton quickly summed up his advice in just five words.... “train to be a plumber”. So when the man who helped create the very neural networks behind AI recommends his children work as plumbers, is it time to worry?
One Irish academic who thinks not is Dr Sara Kieran, associate professor and Director of Executive Education with the University of Limerick’s Kemmy Business School.
She urges caution to those who fear the worst and believes much of what is impacting the workforce is not so much AI but simply a growing “digitisation” of work practice.
This process has in fact been around for years, and is not as new or scary as people might want to think, Ms Kieran said. “This isn't even about AI. Really, this is just digitisation of work, which has been around now for almost 10 years.
“Research I would have done back in 2018 for (the national talent development agency) Skillnet Ireland was looking at digitisation in shared services.
“Shared services are essentially kind of global call centres that may be servicing customers, or they may be servicing the organisation itself. So that would include things like IT, supply chains, HR, finance, that are servicing either customers or or the actual organisation.
“I always think that the easiest analogy is banking because we all understand how our behaviour has changed as consumers of banking. That's not AI. That's just basic digitisation. So anything that is routine and that can be standardised, can be digitised.”
She said people need to understand there are two key issues around jobs — the displacement of jobs, and the replacement.
Displaced means the jobs have gone.
“There's a lot of very standard, routine work that happens in the likes of banks and shared services type centres, call centres that just don't really exist anymore.
“Also, look at the way a lot of IT support work has gone. Our laptops are upgrading, our satellite boxes are upgrading, and we barely even know. A lot of the little bugs in technology are being handled for us without us even knowing.
“So that's been taken away, and as a result, a lot of IT support work and all those routine service transactions are now gone. So there's already been a huge amount of jobs displayed.”
However, the interesting aspect about all this is not so much the number of jobs that have been displaced, she says, but the jobs that have been replaced.
She references the World Economic Forum’s Future of Jobs Reports as being one of the most accurate predictors around work and related trends globally.
The 2025 edition, a deep dive into the future of work over the next five years, predicts roughly 170 million new jobs will be created by global macro trends this decade while 92 million roles will be displaced by these same trends.
The report, based on a survey of more than 1,000 of the largest employers around the world, representing 22 industry “clusters” and more than 14 million workers, says this means there will actually be a net employment increase of 78 million jobs.
The report predicts that frontline job roles will see the largest growth — roles that include those carried out by farmworkers, delivery drivers, construction workers, salespersons, and food processing workers.
Care economy jobs, such as nursing professionals, social work and counselling professionals, and personal care aides are also expected to grow “significantly” over the next five years. So too are the roles of second and third-level teachers.
Technology-related roles are seen as the fastest-growing jobs in percentage terms, such as expert data analysts, fintech engineers, AI and machine learning specialists, and both software and app developers.
The WEF also believes green and energy transition roles, including autonomous and electric vehicle specialists, environmental engineers, and renewable energy engineers also feature within the fastest-growing roles.
The biggest cohort of workers who can expect to see the biggest decline include clerical and secretarial workers, cashiers and ticket clerks, and admin assistants. Other roles expected to shrink include postal clerks, bank tellers and data entry clerks.
On average, workers can expect that two-fifths (39%) of their existing skill sets will be transformed or become outdated over the 2025-2030 period.
However, the WEF says, this measure of “skill instability” has slowed compared to previous editions of the report, from 44% in 2023 and a high point of 57% in 2020 in the wake of the pandemic.
This finding could potentially be due to an increasing share of workers (50%) having completed training, reskilling or upskilling measures, compared to 41% in the 2023 edition.
Analytical thinking remains the most sought after core skill among employers, with seven out of 10 companies considering it as essential in 2025. This is followed by “resilience, flexibility and agility”, along with leadership and social influence.
AI and big data top the list of fastest-growing skills, followed closely by networks and cybersecurity as well as technology literacy.
Complementing these technology-related skills, creative thinking, resilience, flexibility and agility, along with curiosity and lifelong learning, are also expected to continue to rise in importance over the 2025-2030 period.
While global job numbers are projected to grow by 2030, existing and emerging skills differences between growing and declining roles could exacerbate existing skills gaps.
Ms Kieran said:
“We might be losing basic routine, moderately skilled jobs like call centre agents, IT technicians, finance clerks, accountants, clerks because their work is being digitalised.
“But we're creating a lot of jobs, like software engineers, biomedical experts, cyber security experts, financial analysts. If anybody is particularly worried about the job situation, all they have to do is look at employment levels in Ireland.
“People are being employed, and it hasn't been a cliff edge. It’s not that organisations are making swathes of people redundant. There are longer term strategies in play.
“The banking sector has had, for example, voluntary redundancy schemes running for years. So there's natural attrition of people just resigning and retiring, and there are ad hoc, voluntary redundancy schemes, where people are being encouraged to leave early.
“But at the other end of the spectrum they're not recruiting. They're not bringing in swathes of 18-year-olds anymore to do really basic low skills jobs because they don't really exist anymore.”
What might be fueling workers' anxiety is the continuing drip feed of job cuts associated with AI.
Tech giant HP, for example, only recently announced plans to shed 10% of its worldwide workforce globally as it embraces AI.
While this will lead to up to 6,000 jobs being slashed, other tech firms like Google and Amazon have also announced they are reducing staff numbers. But while headlines will grab the attention of anybody worried about their future, caution is advised.
The recent publication of the results of a Massachusetts Institute of Technology (MIT) study into the impact of AI on jobs is a case in point.
If you just read the headlines, you would be understandably terrified. The study appeared to predict that just under 12% of the US jobs market will be replaced by AI.
The study results came from the so-called Iceberg Index, which is based on analysis of 151 million workers across 923 occupations in 3,000 counties covering more than 32,000 skills.
But the study came with some important caveats. MIT researchers actually pointed out some obvious caveats about Project Iceberg. “The Index captures technical exposure, where AI can perform occupational tasks, not displacement outcomes or adoption timelines.
“The Index does not predict job losses, adoption timelines, or net employment effects. Actual workforce impacts depend on firm adoption strategies, worker adaptation, regulatory choices, societal acceptance and broader economic conditions."
Barry O’Sullivan, a professor at University College Cork’s School of Computer Science, is another sceptic of the doom-mongery around AI and job losses.
In fact, he believes AI has been so over-hyped, and so much more money has been invested in it than is showing a return, that the time is ripe for a bust.
“Companies are laying off thousands of people because AI is replacing them, really?” he asks. “Don't believe everything you read, or everything companies will tell you. Besides, where's the actual evidence?
“Also, who are these people? What are they doing? What have they done before? People should look at the detail behind these announcements and the detail just isn’t there.
“You also have to ask who is sending out these messages about massive job cuts because of AI? I believe the people sending out the scary messages that AI is going to be enormously impactful on jobs tend to be companies who have a vested interest in that narrative.
“It’s a narrative that generates business. It also generates investment, and generates public sector investment, because, you know, governments don't want bad things to happen, so they'll invest in these companies.”
He says there is another source for the bad news narrative, and it is the Geoffrey Hintons of this world.
“There's a group of voices who don't want to be caught on the wrong side of something in the future, if it all goes horribly wrong,” he said. “It’s like a dying atheist who decides it is better to hedge their bets than get it wrong.
"Also, you look at the likes of Open AI’s Sam Altman, the man behind ChatGPT. He was before the US Congress a few years back, warning of the dire consequences of the existential threats posed by AI, including the massive impact on jobs.
“Now, if he really believed that, don't you think he'd have gone home and plugged out ChatGPT?”
As to the future, he said people need to remember back to the 1980s when workers were promised that paperless offices — and increased automation — would give everyone so much more free time.
They need to remember because there in lies what he believes is a cautionary tale that contrasts with the “doom-mongering”.
“Despite the promises, the paperless office, the increased automation in the office never did end up giving us more free time,” he said.
“I think we're actually working harder now than we ever ever did, and that's because there's a fundamental law in work called Parkinson's Law.”
A reference to naval historian Northcote Parkinson who, in an essay published in in 1955 said: “Work expands so as to fill the time available for its completion.”
It has various corollaries, including one that states “if you wait until the last minute, it only takes a minute to do” and another that states that “data expands to fill the space available for storage”.
Professor O’Sullivan said:
“So if AI is capable of replacing 50% of the tasks in our jobs, we might actually get the amount of hours we work each day down from 10 to five hours. We’d then get home sooner, and wouldn’t that be a good thing?”
And his prediction of an AI boom turning into a bust has been mirrored by various recent headlines.
Just last month, it was reported that more than $1 trillion had been wiped off US tech giants in what the UK’s called an “AI rout".
It reported that America's eight largest AI-linked stocks had slumped by $1.2trn as investor concerns are mounting over the “sky high valuations” of AI-related stock.
One leading US business website also reported that a number of “top Wall Street CEOs” were warning that “a market pullback could be on the horizon”.
Professor O’Sullivan said:
“I think that they are frantically trying to find a way of recouping the absolutely enormous amounts of money that have gone into it.
"There's investments in the trillions but returns in the millions and there’s just simply no way of earning back the investment. These companies are overvalued and people are in deep.”
Whether or not the bust comes soon or not is anybody's guess, but predictions made by the likes of Geoffrey Hinton and Elon Musk have not come true.
Hinton predicted nine years ago AI would render radiologists obsolete. The Nobel Prize winner famously said “people should stop training radiologists now” because it was “obvious” AI would outperform human radiologists within five years.
That, however, hasn’t happened and neither has the self-driving car revolution promised by Musk. Prof. O’Sullivan points out that there are now more radiologists employed in the Irish health service than ever before.
When he presents his vision for AI in the job market, for example, he draws on the two men for a jokey reference to fantastic AI predictions.
He uses predictions about self-driving cars and radiologists and jokes about how there are more likely to be radiologists stuck in traffic jams driving themselves to work around Cork City than not.
As far as what any worker needs to do, however, he has some basic advice. "It’s very simple. I think anybody who's working in an environment where AI is relevant needs to do some training.
"Just attend training courses and educate yourself. I think people always educate themselves about what's new, regardless of what the doommongers have to say.”
He also points out that the EU AI Act, which came into force in February, states that any organisation deploying or developing AI has to provide training to their employees.
But, for the fearful, and the curious, Dr Kieran suggests a handy website people can search. It’s called willrobotstakemyjob.com.
CONNECT WITH US TODAY
Be the first to know the latest news and updates
