“I say this to everyone in the media world who I talk to,” says Darren Atkins, wrapping up our phone interview: “Please, absolutely do not portray this as a hidden agenda to get rid of staff.”
Atkins is the Chief Technology Office for AI automation at East Suffolk and North Essex NHS Foundation Trust – group of hospitals employing more than 10,000 staff, who serve a quarter of a million people in the South East of England. “If this technology is applied in the wrong way, it can be very threatening,” Atkins says. “Our main priority is to free up time for staff to do the work that they should be doing, rather than the work that has no value.”
Just over a year ago, Atkins led the deployment of virtual workers across his group of NHS hospitals – and according to him, it’s been an unqualified success. Patients are missing fewer appointments and staff are happier. Which, as a result, means happier patients.
This is a story seldom told about the coming wave of high-tech automation in the workplace, with all the associated concerns for automation’s impact on the jobs market.
So, is automation at work a good thing? The story, inevitably, is a little more complex.
Automation of Jobs
“Our economy is facing some fundamental changes,” says Liz Shuler, Secretary-Treasurer of the AFL-CIO, America's biggest federation of labor unions. “We’re already facing questions around the very nature of work.”
This is perhaps the prevailing view of high-tech automation in the workplace – change is coming, it’s happening now, and very few people know what it’s going to look like on the ground. However, one thing that most experts agree upon is that the changes will be significant.
First, let’s consider what we actually mean by automation.
“They’re not robots that sit on the desk,” says Danilo McGarry. “They’re software robots.” Like any burgeoning tech industry, high-tech automation has its detractors and proponents – and McGarry is an expert with a quite particular focus.
“I’ve been running automation programs since 2008, with a heavy focus on financial services industry,” explains McGarry. “I pitched the idea to a major bank to start an automation program in the same month that Lehman Brothers went bankrupt. As half of those working in the industry lost their jobs, automating tasks was a way of maintaining the same level of service to the banks existing clients. I think this gave the bank an edge over its rivals – arguably, one of the many reasons why it shot up the rankings as the financial meltdown unfolded.”
Back in the UK, Atkins’ robots are doing very similar work – filling in patient referral forms, confirming appointments, logging them on NHS internal systems. “The processes that we’ve [automated] already cover work that the member of staff is adding no value to. They’re not applying any logic, or using the skills they’ve spent years learning,” explains Atkins.
“It becomes very mundane and repetitive,” says Atkins. “Staff start making mistakes, they become tired and disconnected, and while I can’t give you any concrete evidence… [the automation of these tasks] has made people seem genuinely happier. They're smiling, they seem more engaged, they seem excited and motivated.”
“You know, they’ve even built themselves little robots out of toilet rolls to put on their desks – they’ve really embraced it.”
Clearly, for Atkins and McGarry, the benefits of automation are two-fold: improved efficiency and happier staff. “This is very tedious manual work. As humans – a beautifully created species – we shouldn’t be doing that kind of stuff anyway,” says McGarry.
“If you pick your use cases for automation carefully, it's a wonderful way of doing very high volume, mundane tasks, but for an unbeatable operating cost,” says McGarry, “If you pick enough use cases to really make a difference to the company's operational costs it can be a very powerful competitive advantage. Most companies can automate at least 20% of their total workload within a space of five or so years. In a multinational that can mean hundreds of millions of dollars in savings”.
But will the upsides of automated work for city-based companies and organizations be felt throughout the world?
Potential Job Loss Through Automation
Perhaps understandably, many people are worried that once companies have a functioning automation program, they will strip the labor force to the bone. Fears about a single machine being able to replace the labor of multiple people are far from new, but there is something different about high tech automation.
McGarry explains that at the moment, most automation focuses around “Robotics Process Automation (RPA), which is robots doing simple tasks.” However, when you combine RPA with things like optical character recognition (OCR) and natural language processing (NLA), “you’re using the bot with some semblance of intelligence, like a human.”
The big change, according to McGarry, is to do with decisions. “When you have big data, you organise and structure that into deep learning, using data science, and then you start on the journey to machine learning – teaching the robot to interpret data – and then you get to start on artificial intelligence and it starts to mimic human decision making.” However, this kind of sophistication is seldom seen. For the moment, at least.
But there’s a catch. While that kind of sophisticated automation will eventually get to workers in developed, service-based economies, so-called “middle income countries” will be particularly vulnerable.
“Shortcomings in basic skills education, among other weaknesses, will severely hamper countries in South and South-East Asia,” a report from The Economist claims. The report, which aggregated a range of metrics to produce an “automation readiness” score, points to a range of middle-income and rapidly developing economies such as Brazil, India, and South Africa as being “unable to capitalize on, and meet the challenges of automation.”
Automation and Manual Work
So far, we’ve spoken about automation and its effects on mostly desk-based or service-based roles. But what about manual work? Surely if a computer can think like a person with enough training, they must be able to carry out menial and manual tasks with ease?
As it turns out, according to Chris Roberts, Head of Industrial Robotics at Cambridge Consultants, they can’t.
“I work with a lot of cutting edge robots, and they can be so stupid,” says Roberts, “If you see a robot going wrong, they look ridiculous. We had a robot who cooks steak and fries, and when it was working, it was brilliant, it was really impressive. But sometimes, it would try and pick the steak up, but instead miss it. It would then carry on doing all the actions – cooking this invisible steak on the hot plate and trying to turn the invisible steak.”
It’s a truism these days to point out that robots are good at things that humans aren’t, and vice versa. Robots demand predictability and regularity, which is why premises like car factories are crammed with automated tech, doing everything from assembling parts on the production line to checking output consistencies in microseconds using machine vision cameras. The agricultural sector, meanwhile, isn’t quite as automated.
Roberts is slightly more complimentary of Mamut, one of Cambridge Consultants' robots. “The genius of Mamut,” he explains, “is that it's an autonomous platform for navigating fields.”
“A lot of our agritech clients have problems or opportunities where they’d like to do something that involves exploring the whole field or set of fields that a farmer owns. You can take a picture of every apple in an orchard to do yield forecasting, or you can see how ripe your crop is. But you can’t do it with a drone because the leaves are in the way – so you need something that can move around the field.”
However, while solutions such as Mamut might be able to extract marginal gains for large agritech firms, the advantages of such a system for most agricultural businesses would be negligible. The initial cost of developing a system like Mamut would be too high for most agricultural businesses. And, of course, when non- or low-skilled manual labor is so cheap, there’s almost no point in developing an autonomous system.
For larger service and information-based businesses, labor – even for entry level roles – is far more expensive, with many large businesses requiring a college-level education to simply get a foot in the door.
And, when the work done by that labor is systematic and ordered, it makes complete business sense to automate those roles. Robots don’t have coffee breaks, arrive at the office hungover, or have petty office feuds. They just do their jobs.
Automation’s Impact on Women
So far, we’ve largely spoken about workers as one monolithic group who will all be subject to the same pressures. But this is neither true, nor helpful for understanding the problems that workers will face.
There are fears about how high-tech automation will affect men and women differently. Some studies claim “that twice as many women as men work in occupations with a high potential for automation.” Some 64% of jobs in these at-risk occupations are believed to be held by women. Statistics such as these are intimidating, and illustrate the disparity between the types of roles women and men are currently working in.
However, according to Henry Parkes, who co-authored the Institute for Public Policy Research’s report looking at the gendered nature of automation, it’s not so straightforward.
“There’s a difference between the technical potential for something to happen and whether it will happen,” explains Parkes. “First of all, it needs to be technically possible. Second, it needs to be socially acceptable that the job in question could be done by a machine – something which is changing over time. And third, whether it’s economically viable – just because you can do something with a robot, that doesn’t mean it makes business sense.”
It’s not all that simple, though. “We know that women are less likely to hold STEM qualifications, so in one respect they’re more vulnerable to automation, as they’re not positioned to take advantage of the new opportunities that might become available,” says Parkes.
“But, for instance, look at the care sector,” says Parkes. “There might be a role for some automation, but there’s still going to be a very important human aspect to it all. There will be automation, but people will be working with machines to improve their service.”
In fact, according to McKinsey, women are moving out of highly automatable roles and into less automatable (and often better paid) jobs. Between 2001 and 2018, 450,000 female workers in Britain upped and left their jobs in secretarial and administrative roles. Meanwhile, some 1 million women started working in less automatable roles in project management, legal and teaching professions, and medical practice. Similar trends have also been observed in the US, where healthcare is set to be the only sector of the US economy which will see growth in physical and manual skills between now and 2030.
Of course, while some might be heartened to hear that industries like the care and medical sectors are growing and in need of distinctly human workers, others may fear that we are in danger of returning to a distinctly gendered understanding of work – men do coding, while women do caring.
“Women are woefully underrepresented in tech,” says Parkes, “and we might find ourselves widening the gender pay gap – which has been slowly shrinking – if all the new highly paid jobs are in tech.”
Unfortunately, fixing that problem will take years, if not generations of societal change to get women into tech and men into people-facing roles. That doesn’t mean it shouldn’t be tried, nor that it is impossible. But it will be a harder problem to fix.
Are We Being Protected from Automation Job Loss?
High tech automation may be a relatively new phenomenon, but it’s been on the radar of top tech company executives for a while. Bill Gates has been giving interviews about the threat posed by AI and automation to workers for the past five years, at least. However, government officials and politicians are also aware of the potential impacts.
“Automation is a serious threat to the jobs of many workers,” says Sharon Graham, Executive Officer of the British labor union Unite. “We are developing a political and industrial strategy to build a future that works.”
However, it's not just labor unions that are trying to pressure governments into action. In its Future of Work report, market research company Forrester slammed governments’ readiness to address the issue: “We already see many in power who don’t understand today’s digital world. Governments will need to move faster than most to catch up to a digital-automation dynamic that can threaten or empower their jurisdictions.”
Some governments may be planning for the social impact of an automated future, but they’re typically far less prepared than the companies who stand to profit from high tech automation. However, those in power may soon be faced with a rude awakening.
“Companies are effectively going to be earning more, because they’re going to be saving more money, with less corporate tax revenues. But fewer people are going to be earning more. So, that’s going to leave a hole in government budgets,” explains McGarry. “Should governments be taxing robots? Should governments consider the robots that complete tasks for companies as humans? The whole tax system needs to be completely revamped in order for that to be effective.”
Of course, government budgets aren’t the only heads on the chopping block. As Liz Shuler from the AFL-CIO said earlier, we’re facing questions about the very nature of work.
After his work with the bank, McGarry explained: “There were fewer and fewer bodies visible in the office, with more people working from home. People were being converted to contractors, doing three days a week instead of five. Eventually, towards my final year, I’d say roughly 20% of the operational people were just on call, in case the bank needed them.”
“That’s what’s going to happen to the wider workforce – people will go from full-time employment, with pensions, benefits, and healthcare, to being consultants that work part-time, to being people on call. That’s a huge shift in how we work and the dynamics of working.”
This change would require a complete re-think in the way workers, governments, and companies treat careers and training. “Intelligent automation is expected to boost the importance of both education related to STEM and of so-called soft skills, which allow workers to trade on their uniquely human capabilities,” explained the economist. “However, little has been done to prepare future workers through school curricula – or, just as importantly, teacher training.”
An Automated Workplace – What Happens Next?
We’d be lying to you if we said that anyone truly knows what the workplace will look like in 10 or 15 years time.
It’s likely that the change will be gradual. Robots won’t take your job overnight, and there’s a fair chance that one might not take your job at all. However, there might be aspects of your job that are passed onto a robot. It’s perhaps more likely that the jobs of the future involve commanding groups of robots to do work, analyzing and optimizing their performance, and fixing them when they break down.
It is clear, however, that high-tech automation will become a fixture in every workplace. From care homes and trading floors to news desks, automated machine intelligence will be carrying out our tasks sooner than we imagine. Perhaps, even the way we work will change beyond recognition.
As the populations around the world get older, the health and care sector will grow. As companies begin to automate more functions, workers will be controlling lots of separate functions, rather than focusing on one specific area.
But, as with any period of upheaval, some are perhaps rightfully worried about the potential for wealth concentration and worker exploitation.
“New technology isn’t inherently good or bad,” says the AFL-CIO’s Shuler. “It’s a tool, and we want working people to have a say in how that tool is used. When that technology creates new wealth and prosperity, we want to ensure that everyone shares in it – not just a few billionaires and corporate executives.”
“We need a radical response to the new realities of the labor market, both in the public and in the private sector,” says Unite’s Graham. “This challenge is beyond the old politics – tinkering at the edges is not a solution. New technology is going to generate a lot of wealth. We need to make sure this wealth is used to protect jobs, and to improve the lives of workers and their families.”
Some would say, however, that the responses of labor unions are not only predictable, but unhelpful. New technology is the free market at work, with the invisible hand guiding technological developments which have the potential for driven individuals to create wealth and opportunity for others.
However, as our interview drew to a close, Danilo McGarry issued a stark warning about the way work might change.
“It’s all wrong, capitalism and commercialism, the way it’s been shaped to generate money. I really think that the changes that automation will bring will make people question what makes them happy in life.”
“And, you know, they might realise it’s not buying stuff, it’s not materialism – it’s minimalism and humanism. Coming from an ex-banker, that’s pretty shocking.”
If that is the next step, we’re all in for one hell of a ride.