Dr Mark Nasila | Chief Analytics Officer | Chief Risk Office | FNB | mail me |
The era of emerging technologies is changing everything in our lives. It’s changing the jobs landscape and how we work, but also how we interact with one another, how we are entertained, and how we learn.
At the same time, businesses face changing customer expectations, and governments must contend with new demands from their citizenry. In short, we are in the midst of the 4th Industrial Revolution (4IR), and it is changing everything.
On the cusp of becoming mainstream
We are at an inflection point where emerging technologies are on the cusp of becoming mainstream. Services like ChatGPT and Midjourney — which use artificial intelligence (AI) to create text and images respectively — are upending how we think about education, creativity, art, and culture.
So are other AI-powered services like English Language Speaking Assistant (ELSA), which can understand non-native English speakers’ accents, ChatSonic, Replika, Socratic, and Google’s forthcoming ChatGPT rival, Bard.
AI diagnostics are changing the medical landscape, and now AI and data are also being harnessed to power China’s mass surveillance apparatus. At the same time, global agriculture is using AI to improve yields and predict disease outbreaks or weather anomalies. Another emerging technology, virtual reality, is being used to augment training, assist architects, quantity surveyors, and builders, and even enable doctors to perform or guide surgery remotely.
Closer to home, AI is transforming accounting services by improving document processing, automating authorisation, flagging fraudulent or suspicious activity, and providing deeper insights, or existing ones but far more rapidly than has previously been possible.
AI also enhances accounting processes, from streamlining procurement and purchasing to invoicing, purchase orders, expense reports, and other essential — but often time-consuming — activities.
What is AI?
But what is AI? It’s a type of advanced technology that mimics human intelligence. It uses tools like machine learning to recognise patterns in datasets, and deep learning to be able to apply what it learns to fresh instances.
McKinsey predicts the AI industry will generate more than $16 trillion in value across all industries in coming years, with the United States and China expected to add $10.7 trillion of that, and 14.5% and 26% to their respective GDPs from AI-powered activities. This dominance from the two superpowers is not surprising.
The US and China are the biggest spenders on AI technology, and the countries producing the most AI-related research. The US has the most AI-focused startups, while China has a population that’s very enthusiastic about and receptive to the technology. A study from Deloitte found that 77% of Chinese business leaders believe AI will transform their businesses in the next three years. Meanwhile, 69% of US business leaders believe AI is “very or critically important” to their companies’ success.
Why do companies need AI?
Purely from a business perspective, AI has obvious commercial benefits. It can make operations more efficient by identifying problems, automating certain activities away, improving data analytics, and freeing up skilled workers to concentrate on decision-making, each of which brings down costs.
AI can also have humanitarian benefits. It can improve workers’ experience by sparing them from rote or laborious tasks, and it can also take over tasks where human error can creep in.
It’s also important to realise that this sort of AI is here already – it’s not some far-fetched fantasy that’s only due to arrive years down the line. Many businesses are using it already, which is why it’s so important companies embrace it and begin to understand its potential as soon as possible. Not doing so means they risk falling behind their rivals, but also that they’ll lose customers to businesses better positioned to meet those customers’ needs.
By successfully deploying AI, businesses can make faster and better decisions, they can predict customer preferences, and offer more bespoke services to them. Businesses that embrace AI can also more effectively turn data into actionable insights and can maximise revenue by identifying and optimising sales opportunities, all while saving time and money. But doing so requires buy-in from across the business, and a willingness to go through the process of integrating AI into multiple aspects of the enterprise.
A study by Infosys found the primary driver for using AI in business is the competitive advantage it can offer. Thereafter, other motivators included solving particular business, operations, or technical problems, demand from customers, an internal experiment that showed the potential use cases, or simply an executive-led decision that leads to its rollout.
Automating the repetitive, bolstering the creative
Kai-Fu Lee, the author of the 2018 book AI Superpowers, says he expects a growing number of repetitive jobs will be automated by AI in the coming years. These include repetitive talks like customer support, factory line work, and dishwashing – which he expects will begin to vanish within five years – and routine ones like driving trucks or providing access and security services, which he expects to be taken over by AI in the next decade.
Lee creates four categories for work: those with optimisation vs. creativity, and low vs. high empathy requirements. AI excels at optimisation and low empathy jobs, while humans excel when jobs require creativity and high levels of empathy or compassion. For example, AI is extremely adept at repetitive roles and wrestling with enormous data sets jobs. It’s good at writing articles on numbers-heavy sporting events or year-end financial reports, but not at writing columns debating ethical challenges or a book review.
However, that’s not to say there’s no room for AI outside of the optimisation/low empathy realm, or that there’s no place for humans outside high creativity/empathy spaces. Instead, there’s lots of room for overlap. For example, AI can help a columnist with research for a story, or it can help a doctor with a diagnosis by comparing a patient’s symptoms to thousands of other medical records in an instant. AI can help detect the most urgent crisis hotline calls and then route them to the human best suited to offer support in the circumstances.
What does it mean for careers? It means the chance to find more meaningful roles, and when aiming to reposition those people whose roles may become obsolete, we should focus on reskilling them for more humanity-focused roles, rather than for other soon-to-be-automated ones.
It also means some careers will be improved. For instance, AI can help personal trainers make custom plans for their clients taking multiple metrics into account and adjusting as those metrics change. Beauty consultants will use AI to manage bookings. Teachers and remote tutors will use it to help plan curricula and adapt them in nearly real-time to individual students’ needs. Wedding planners will use AI to manage vendors while tour guides will use it to create itineraries that take into account clients’ previous trips, activities, and preferences.
If we don’t allow technology to take over routine tasks, we limit the potential scope of our humanity, or the potential to live the sort of lives we’ll later be satisfied with. Where early AI of the sort that beat Gary Kasparov at chess used sheer computational might, today’s AI has moved beyond brute force calculations and can understand circumstance and context and adapt on the fly. That means where it can’t be used to automate careers, it may often still be able to augment them.
Automation removes humans from tasks, augmentation empowers them to do them better. Where jobs require creativity or empathy, contemporary AI solutions have the ability to enhance our abilities to deliver those things, by driving complex functions that require both, like the work done by concierges, social workers, and marketing directors, for instance. AI can liberate people to do more human things by making them more effective or efficient, or by enabling them to deploy their expertise or their ability to innovate.
What about the pessimists?
There are, inevitably, those who argue that AI isn’t yet good enough or that we as a society aren’t ready to use it effectively. They point to problems with biases creeping into AI, like the disastrous chatbot Tay that rapidly began spouting hateful rhetoric. They also point to stories like Cambridge Analytica abusing Facebook’s tools to sway on-the-fence voters in the US, and the failure of Boeing’s 737 Max’s automation systems that led to aircraft crashes.
But these instances don’t demonstrate it’s too early for AI. Instead, they point to how we should be thinking about the AI development lifecycle and the importance of putting the right checks, measures, and guardrails in place to ensure AI develops in a way that benefits humanity and improves our lives rather than jeopardising them.
In the planning phases of the AI lifecycle, it’s important to choose a problem of scale, find the right technology and tech provider to address it, design the solution, assess it, iterate, deploy it, and then be willing and able to change it as required. Technology is an extension of humanity, so the onus is on us to ensure it reflects the parts of our humanity we want to amplify.
As author Gwen Moran argues, “Al has a bias problem, and only humans can make it a thing of the past.”
The power of design thinking
One way of humanising tech is to develop design thinking (DT) skills and following related best practices.
Design thinking places the user at the centre of product creation. It’s based on empathy as a driving force for innovation. Design-driven companies have consistently outperformed traditional ones, because design thinking as a skill is helping employees become more human-centric and better meet human needs in the process. If employers can focus on DT skills, they can become more humanist in the process, because DT humanises data science processes.
Tim Brown, CEO of IDEO, an international design and consulting firm, explains: “Design thinking is a human-centred approach to innovation that draws from the designer’s toolkit to integrate the needs of people, the possibilities of technology, and the requirements for business success.”
The future will likely be filled with human-machine partnerships, and assuming the right approach is used to get there, that’s good news for humanity. Because AI is poised to provide assurance and hope to humanity, freeing us from the mundane while helping us perform better at the things we’re best at. No company will succeed without a human-centric focus, and AI is creating precisely the sorts of new jobs that demand compassion and empathy.
