AI in the workplace: what should CEOs really be thinking about?
There are a number of important factors leaders should consider when it comes to getting the most out of AI and machine learning at work
Most people would be familiar with artificial intelligence (AI) technologies such as Apple’s Siri, Amazon’s Alexa or maybe even IBM’s Watson. But in the world of work, AI and machine learning are reshaping – and in some cases, completely transforming – how work is done. From Rio Tinto’s autonomous iron ore haulage trucks and Amazon’s robot-driven warehouses, through to Tesla's autopilot self-driving cars and Google Cloud's Healthcare Data Engine which generates big data-driven healthcare insights, artificial intelligence is being deployed in new and important ways.
As with most new advancements in technology, there are upsides and downsides – both of which need to be carefully considered by organisations looking to utilise AI. One of the commonly cited issues associated with use of AI systems in the world of work is job losses. Many such fears can be traced to a 2013 University of Oxford study which predicted that almost half of all the jobs in the US were at risk of automation within “a decade or two”.
However, the conclusions of such studies need to be treated with caution, according to Toby Walsh, a Scientia Professor of Artificial Intelligence at UNSW Sydney. In a submission to a senate committee on the future of work, Prof. Walsh said there are many factors typically not considered by such studies. And while artificial intelligence, machine learning and other types of AI technologies are significantly disrupting the world of work through automation, there are new jobs being created by such technologies, which not only require new skills in the field of AI and computer sciences on the part of human workers, but can augment and enhance the life of work in valuable ways. And this is where there is a significant upside for both organisations and individuals.
“I think an important point to make is that artificial intelligence is generally a net creator of jobs, rather than a destroyer of jobs,” says Aaron McEwan, Adjunct Senior Lecturer in the School of Population Health at UNSW Sydney and VP, Research & Advisory for Gartner. “That’s generally because artificial intelligence doesn’t really take over an entire job; it just takes over parts of almost everybody’s work.”
McKinsey AI research, for example, found that about 30 per cent of the activities in 60 per cent of all occupations could be automated through AI work. And while nearly all occupations will be affected by automation, only about 5 percent of occupations could be fully automated by currently demonstrated data science technologies.
Mr McEwan explained AI technology initiatives tend to chip away at less complex, more repeatable tasks, which leads to work that is more cognitively taxing, or potentially more complex and ambiguous. “And it’s certainly more social and creative. So, ironically, it creates work that is more human and much more reliant on what humans can do uniquely that machine learning algorithms can’t do,” he said.
The people skills which AI can’t replace
Some of the early adopters of AI have learned some important lessons about its impact on both workers and customers. A classic example can be found in the insurance industry, in which many businesses use a combination of machine learning and AI-driven algorithms and chatbots for specific tasks ranging from responding to customer queries and assessing and processing claims, through to underwriting and fraud detection.
However, a significant degree of sensitivity is required in working with some customers – which goes beyond the natural language processing and speech recognition capabilities of AI-enabled chatbots. “Take the once-in-a-generation floods that happened in Lismore,” said Mr McEwan. “People might be calling up because they have just lost their home in a natural disaster. They may have even lost members of their family, or they may have lost their pets. We’re talking about people that are calling up in a distressed state with complex needs. What’s required of a call centre operator today is not just this simple transactional interaction with a human being; it’s actually a deeply emotional and complex one. And that requires a much higher level of skill than what artificial intelligence is capable of.”
Read more: How to avoid the ethical pitfalls of artificial intelligence and machine learning
Another example can be found in the fast-food industry. Technology has made the process of ordering quicker (and maybe easier) through the deployment of self-serve touchscreens within McDonald’s outlets, for example. While most customers would be relatively happy to use touchscreens or place a drive-through order, customers who are unhappy or angry usually want to speak with a human being.
“So what does that mean? A poor 16-year-old kid is probably dealing with customers that are angry, certainly in the middle of the pandemic. Or it might be somebody stacking shelves at Woolworths. There’s a really good chance that, even if it is a somewhat semi-skilled or low-skilled job, it actually requires a much higher degree of interpersonal skills and conflict management,” said Mr McEwan. “There’s a reasonable argument that your average fast-food operator may have to do more de-escalation than a police officer today. I say that tongue in cheek, but I wouldn’t be surprised if it’s on par these days.”
Read more: AI: friend or foe? (and what business leaders need to know)
Navigating to the future of work with AI capabilities
AI solutions are proving to be a two-edged sword in the workplace. While AI can assist workers by taking on more repetitive, mundane tasks, there is an emerging body of research that suggests AI is also leading to an intensification of work. “Because it increases in complexity, there are also arguments that it will become more meaningful. You could say that’s been a problem over the past 50 years or so, that people are losing the meaning in work. But there is a kind of insidious downside to this, which not a lot of attention is being paid to this,” said Mr McEwan.
“And it’s to do with the removal of the simple and less complex parts of our day. If you just keep pushing the value bar up into more and more complex, emotional and challenging work, it’s going to become very exhausting very quickly. We’re already in the middle of a mental health pandemic at work, and people are exhausted, burnt out and struggling with workloads. So the downside of AI is that while we might end up with more meaningful jobs, they’re also going to be more difficult and taxing. And our current approaches to the way that we design work are not really built for that level of complexity or effort that’s required.”
As a result, Mr McEwan said the future of work is going to be similar to what elite athletes experience, with intense bursts of work and periods of rest, reflection or routine work (that hasn’t been taken on by AI) in between. “An analogy I often share is that you can get a bunch of disgruntled, underfed, malnourished and poorly treated serfs to build you a temple. But you can’t get them to paint a masterpiece on its ceiling. And the future of work for humans is painting the masterpiece, while the robots will build the temples. If we really want a workforce that can be that creative, look at Michelangelo. He wasn’t working like a slave, thanks to the support of wealthy benefactors that were making his life relatively easy to produce quality work. He painted the Sistine Chapel between the hours of 5am and 10am, and then he took the rest of the day off,” said Mr McEwan.
Applications of AI for leaders
While it is well established that AI is already making a significant difference through automating certain business processes and lower-level tasks for workers usually at the bottom of the organisational pyramid, some of the biggest changes brought on by AI will be felt at the very top of business. And organisational responses to the pandemic have only accelerated these changes, with a shift to remote and hybrid working arrangements, according to Mr McEwan.
“In a world where people don’t work in offices, the responsibility of managers becomes less about directing resources – kind of like air traffic control. In this world, the real job of managers is looking after your people and making sure they have everything they need. Are they in the right place to do their best work? Are they being cared for as, as people (not just workers)? This is important for the supervisory middle management level,” he said.
The deployment of resources is usually the responsibility of those who are the next level up in the organisation, and as AI develops in sophistication it will take on an increasingly important role with real-time decision-making through deep learning, beyond the problem-solving abilities of what the human brain and human intelligence are traditionally capable of. “AI will do a better job of understanding what your unique strengths are and how they can be applied to work that needs to be done, which is something that we’ve historically left up to leaders when it comes to moving their people around effectively,” he said.
Subscribe to BusinessThink for the latest research, analysis and insights from UNSW Business School
And at the very highest levels of organisations, big decisions around commercial drivers such as productivity, revenue and profit will be augmented by AI through harnessing and leveraging the power of big data sets. The world of work is increasingly complex, and leaders such as C-suite executives and board members need to take into account a broad range of considerations when it comes to strategic decisions.
“Leaders will need to ask questions such as ‘What is our responsibility not just to our shareholders, but our employees?’ and ‘What is our responsibility to the country and other countries we operate in, the societies we contribute to, and more broadly to the environment and the planet?’ It will be important to view decisions through the lens of ethics, so leaders will be more focused on much higher-level human decisions around factors such as sustainability, innovation and wellbeing that require wisdom and perspective,” said Mr McEwan.