Commentary: Artificial intelligence and automation would actually benefit Singapore
Machine learning and other digital innovations can help Singapore businesses and workers increase productivity and maintain their competitive edge, says NUS Professor Kenneth Huang.
SINGAPORE: Now that the General Election is over, it is time for Singapore to refocus on the big challenge of creating jobs to tide citizens over a pandemic and double down on digitalisation for the long term.
Much has been said about the concerns people have about livelihoods, with suggestions to safeguard and improve the prospects of jobs for Singaporeans.
Yet disruption is not new to Singapore. History has witnessed how Singapore has upskilled its workforce through computerisation and automation in the 1980s.
Singapore businesses and workers are no strangers to the need to adapt to new technological changes.
Now, Digital Ambassador Corps have been deployed to help small businesses and senior citizens learn and apply technology.
With every change comes hesitance, even resistance. In the push for a Smart Nation, this resistance may come from a fear of the unknown. Reports of artificial Intelligence (AI) and digital technologies cannibalising jobs do not help either.
However, Singapore is in a unique situation. With a small and ageing workforce, Singapore has to tap on AI and automation to preserve its competitive advantage over other economies.
READ: Commentary: Digitalisation push in Fortitude Budget is the right move. Much will depend on whether SMEs embrace this opportunity
A COUNTRY INCREASINGLY POWERED BY MORE ARTIFICIAL INTELLIGENCE
Digital technologies and AI (including machine learning, computer vision and natural language processing) can boost efficiencies, performance and productivity in various ways.
It is these advanced technologies that help e-commerce retailers like Lazada sell more by analysing massive amount of data, learning customer preferences and providing targeted products to be displayed online for the customers.
In engineering and aviation, AI has been used to increase the performance of gas turbine engines, such as finding an optimal way to increase thrust and decrease fuel consumptions.
In the long term, the savings on fuel could be passed to the passengers. Such performance improvements cannot usually be attained using traditional models.
In logistics in Singapore and around the world, AI has also been utilised to predict traffic patterns and route conditions. For companies like Grab, the use of AI has enabled drivers to complete as many jobs as possible in the shortest amount of time.
Grab also uses natural language processing methods to address customer feedback and enable users to find the services they need with greater ease.
AI is also extensively used in the development of autonomous vehicles like the National University of Singapore (NUS) autonomous shuttle at its Kent Ridge campus.
In healthcare, AI has been employed to optimise hospital management and processes like managing a large number of patient beds in the case of Tan Tock Seng Hospital. Predictive analytics can help optimise hospital bed assignment decisions by predicting when patients will be discharged to make more beds available.
AI will be an integral part of Singapore’s healthcare system to help doctors make better decisions and design early intervention programmes and improved care pathways for patients using predictive modelling.
One application of machine learning is “precision medicine” where AI can help predict what treatment protocols are most feasible and with higher success rate on a patient based on various patient characteristics and the treatment context.
Another example is robotic surgery (like the da Vinci Surgical System used in Gleneagles Hospital Singapore) which can help surgeons improve their ability to perform precise and minimally invasive incisions and surgeries. Important decisions are still made by human surgeons.
In educational applications and tools, AI has helped the development of skills and testing systems and allows the adjustment of learning based on differentiating students’ needs in Institutes of Higher Learning in Singapore.
Students can thus enjoy more customised testing and learning tailored to the specific needs and ability level of each student.
JOBS ARE CHANGING
In areas where AI and digital technologies improve businesses significantly, the nature of jobs has changed.
Certain jobs like routine clerical work may be reduced while the employment rates for professionals and those in the service sectors have increased. Understanding what tasks AI is suited or not suited for will be a business priority for firms. Singapore’s learning, retraining and upskilling efforts must take full advantage of the AI era.
Prior research has shown AI is suited to perform tasks that provide clear feedback with definable goals and metrics. AI is also efficient at recognising associations based on empirical and statistical data.
AI can help improve traffic volume and flow in metropolitan areas like Singapore, New York or London using pre-defined performance and congestion measures at the system level by analysing large amounts of traffic data.
READ: Commentary: The revival of the digital economy – building citizen confidence as Smart Nation momentum picks up
On the other hand, AI is not so good at unstructured tasks and reasoning, especially based on background information that is previously unknown to the computer.
This is why AI (or machine learning) can be used to spot irregular heartbeat from scans and detect diseases from medical imaging, but it cannot explain as well as doctors how and why one is diagnosed with a certain disease.
In other words, the interpretation of the causes and severity of these diseases and their linkages to other diseases are much more difficult for AI to ascertain. AI also does not perform well when the tasks to be learned change quickly.
Humans do much better at interpreting data and drawing inferences even when the tasks evolve over time.
YOUR JOB MAY REMAIN BUT SOME TASKS COULD BE OUTSOURCED TO AI
In light of the above understanding, how we should we adjust, retrain or upskill the valuable human resource we have in Singapore to prepare for the new paradigm involving AI and digital technologies?
We understand that most jobs have many interrelated tasks. People say the jobs AI could likely replace include telemarketing, receptionists, computer support specialists (think chatbots used by banks like OCBC) and market research analysts.
However, it doesn’t mean these jobs will disappear entirely. AI is weak on relatively unstructured, creative tasks and those involving emotional intelligence.
The focus of the training or upskilling of such roles should be on these areas. Upskilling courses can cover developing strategies in branding, designing and marketing.
Use AI to gather your data, but use humans to develop business and innovation strategies and design marketing campaigns based on understanding those data.
People and leadership skills will continue to be important, yet another area that AI currently does not fill the void. The expertise in asking interesting questions and looking for new and innovative solutions, which is required in researchers or entrepreneurs, will also be deemed more valuable.
READ: Commentary: This COVID-19 outbreak, corporate leaders should acknowledge they don’t have all the answers
The age of AI and digital technologies is already here. It is clear they can and probably should be applied to different industries and have the potential to significantly improve productivity.
In the process, they will transform our work and lives. While some jobs may be replaced, many other job and career opportunities will be created.
Singapore has the infrastructure, talents and resources to take advantage of the benefits brought about by the AI revolution.
With national emphasis on innovation and Industry 4.0, as well as additional resources and upskilling opportunities, this could yet be another pivotal point for Singapore to create and deliver value in a competitive global arena.
Dr Kenneth G Huang is an Associate Professor with the Department of Strategy & Policy at National University of Singapore (NUS) Business School and the Department of Industrial Systems Engineering & Management at NUS. The opinions expressed are those of the writer and do not represent the views and opinions of NUS.